100% found this document useful (6 votes)
20 views

Intelligent Computing Proceedings of the 2020 Computing Conference Volume 3 Kohei Arai download

The document presents the proceedings of the 2020 Computing Conference, edited by Kohei Arai, Supriya Kapoor, and Rahul Bhatia, featuring a collection of research papers across various topics in intelligent computing. It includes contributions from over 500 submissions, with 160 selected for publication, covering areas such as machine learning, security, and intelligent systems. The series aims to rapidly disseminate significant developments in intelligent systems and computing through conference proceedings and is indexed in major databases for wider accessibility.

Uploaded by

atikairvinhj
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (6 votes)
20 views

Intelligent Computing Proceedings of the 2020 Computing Conference Volume 3 Kohei Arai download

The document presents the proceedings of the 2020 Computing Conference, edited by Kohei Arai, Supriya Kapoor, and Rahul Bhatia, featuring a collection of research papers across various topics in intelligent computing. It includes contributions from over 500 submissions, with 160 selected for publication, covering areas such as machine learning, security, and intelligent systems. The series aims to rapidly disseminate significant developments in intelligent systems and computing through conference proceedings and is indexed in major databases for wider accessibility.

Uploaded by

atikairvinhj
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 60

Intelligent Computing Proceedings of the 2020

Computing Conference Volume 3 Kohei Arai


download

https://textbookfull.com/product/intelligent-computing-
proceedings-of-the-2020-computing-conference-volume-3-kohei-arai/

Download more ebook from https://textbookfull.com


We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Intelligent Computing Proceedings of the 2020 Computing


Conference Volume 2 Kohei Arai

https://textbookfull.com/product/intelligent-computing-
proceedings-of-the-2020-computing-conference-volume-2-kohei-arai/

Intelligent Computing Proceedings of the 2020 Computing


Conference Volume 1 Kohei Arai

https://textbookfull.com/product/intelligent-computing-
proceedings-of-the-2020-computing-conference-volume-1-kohei-arai/

Intelligent Computing: Proceedings of the 2018


Computing Conference, Volume 2 Kohei Arai

https://textbookfull.com/product/intelligent-computing-
proceedings-of-the-2018-computing-conference-volume-2-kohei-arai/

Intelligent Systems and Applications: Proceedings of


the 2020 Intelligent Systems Conference (IntelliSys)
Volume 3 Kohei Arai

https://textbookfull.com/product/intelligent-systems-and-
applications-proceedings-of-the-2020-intelligent-systems-
conference-intellisys-volume-3-kohei-arai/
Intelligent Systems and Applications: Proceedings of
the 2020 Intelligent Systems Conference (IntelliSys)
Volume 2 Kohei Arai

https://textbookfull.com/product/intelligent-systems-and-
applications-proceedings-of-the-2020-intelligent-systems-
conference-intellisys-volume-2-kohei-arai/

Proceedings of the Future Technologies Conference (FTC)


2020, Volume 1 Kohei Arai

https://textbookfull.com/product/proceedings-of-the-future-
technologies-conference-ftc-2020-volume-1-kohei-arai/

Proceedings of the Future Technologies Conference FTC


2018 Volume 2 Kohei Arai

https://textbookfull.com/product/proceedings-of-the-future-
technologies-conference-ftc-2018-volume-2-kohei-arai/

Proceedings of the Future Technologies Conference FTC


2018 Volume 1 Kohei Arai

https://textbookfull.com/product/proceedings-of-the-future-
technologies-conference-ftc-2018-volume-1-kohei-arai/

Advances in Computer Vision: Proceedings of the 2019


Computer Vision Conference (CVC), Volume 1 Kohei Arai

https://textbookfull.com/product/advances-in-computer-vision-
proceedings-of-the-2019-computer-vision-conference-cvc-
volume-1-kohei-arai/
Advances in Intelligent Systems and Computing 1230

Kohei Arai
Supriya Kapoor
Rahul Bhatia Editors

Intelligent
Computing
Proceedings of the 2020 Computing
Conference, Volume 3
Advances in Intelligent Systems and Computing

Volume 1230

Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland

Advisory Editors
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing,
Universidad Central de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, School of Computer Science and Electronic Engineering,
University of Essex, Colchester, UK
László T. Kóczy, Department of Automation, Széchenyi István University,
Gyor, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas
at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao
Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information Technology,
University of Technology Sydney, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute
of Technology, Tijuana, Mexico
Nadia Nedjah, Department of Electronics Engineering, University of Rio de Janeiro,
Rio de Janeiro, Brazil
Ngoc Thanh Nguyen , Faculty of Computer Science and Management,
Wrocław University of Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering,
The Chinese University of Hong Kong, Shatin, Hong Kong
The series “Advances in Intelligent Systems and Computing” contains publications
on theory, applications, and design methods of Intelligent Systems and Intelligent
Computing. Virtually all disciplines such as engineering, natural sciences, computer
and information science, ICT, economics, business, e-commerce, environment,
healthcare, life science are covered. The list of topics spans all the areas of modern
intelligent systems and computing such as: computational intelligence, soft comput-
ing including neural networks, fuzzy systems, evolutionary computing and the fusion
of these paradigms, social intelligence, ambient intelligence, computational neuro-
science, artificial life, virtual worlds and society, cognitive science and systems,
Perception and Vision, DNA and immune based systems, self-organizing and
adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics
including human-machine teaming, knowledge-based paradigms, learning para-
digms, machine ethics, intelligent data analysis, knowledge management, intelligent
agents, intelligent decision making and support, intelligent network security, trust
management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are
primarily proceedings of important conferences, symposia and congresses. They
cover significant recent developments in the field, both of a foundational and
applicable character. An important characteristic feature of the series is the short
publication time and world-wide distribution. This permits a rapid and broad
dissemination of research results.
** Indexing: The books of this series are submitted to ISI Proceedings,
EI-Compendex, DBLP, SCOPUS, Google Scholar and Springerlink **

More information about this series at http://www.springer.com/series/11156


Kohei Arai Supriya Kapoor
• •

Rahul Bhatia
Editors

Intelligent Computing
Proceedings of the 2020 Computing
Conference, Volume 3

123
Editors
Kohei Arai Supriya Kapoor
Faculty of Science and Engineering The Science and Information
Saga University (SAI) Organization
Saga, Japan Bradford, West Yorkshire, UK

Rahul Bhatia
The Science and Information
(SAI) Organization
Bradford, West Yorkshire, UK

ISSN 2194-5357 ISSN 2194-5365 (electronic)


Advances in Intelligent Systems and Computing
ISBN 978-3-030-52242-1 ISBN 978-3-030-52243-8 (eBook)
https://doi.org/10.1007/978-3-030-52243-8
© Springer Nature Switzerland AG 2020
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Editor’s Preface

On behalf of the Committee, we welcome you to the Computing Conference 2020.


The aim of this conference is to give a platform to researchers with fundamental
contributions and to be a premier venue for industry practitioners to share and
report on up-to-the-minute innovations and developments, to summarize the state
of the art and to exchange ideas and advances in all aspects of computer sciences
and its applications.
For this edition of the conference, we received 514 submissions from 50+
countries around the world. These submissions underwent a double-blind peer
review process. Of those 514 submissions, 160 submissions (including 15 posters)
have been selected to be included in this proceedings. The published proceedings
has been divided into three volumes covering a wide range of conference tracks,
such as technology trends, computing, intelligent systems, machine vision, security,
communication, electronics and e-learning to name a few. In addition to the con-
tributed papers, the conference program included inspiring keynote talks. Their
talks were anticipated to pique the interest of the entire computing audience by their
thought-provoking claims which were streamed live during the conferences. Also,
the authors had very professionally presented their research papers which were
viewed by a large international audience online. All this digital content engaged
significant contemplation and discussions amongst all participants.
Deep appreciation goes to the keynote speakers for sharing their knowledge and
expertise with us and to all the authors who have spent the time and effort to
contribute significantly to this conference. We are also indebted to the Organizing
Committee for their great efforts in ensuring the successful implementation of the
conference. In particular, we would like to thank the Technical Committee for their
constructive and enlightening reviews on the manuscripts in the limited timescale.
We hope that all the participants and the interested readers benefit scientifically
from this book and find it stimulating in the process. We are pleased to present the
proceedings of this conference as its published record.

v
vi Editor’s Preface

Hope to see you in 2021, in our next Computing Conference, with the same
amplitude, focus and determination.

Kohei Arai
Contents

Preventing Neural Network Weight Stealing


via Network Obfuscation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Kálmán Szentannai, Jalal Al-Afandi, and András Horváth
Applications of Z-Numbers and Neural Networks in Engineering . . . . . 12
Raheleh Jafari, Sina Razvarz, and Alexander Gegov
5G-FOG: Freezing of Gait Identification in Multi-class Softmax
Neural Network Exploiting 5G Spectrum . . . . . . . . . . . . . . . . . . . . . . . . 26
Jan Sher Khan, Ahsen Tahir, Jawad Ahmad, Syed Aziz Shah,
Qammer H. Abbasi, Gordon Russell, and William Buchanan
Adaptive Blending Units: Trainable Activation Functions for Deep
Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Leon René Sütfeld, Flemming Brieger, Holger Finger, Sonja Füllhase,
and Gordon Pipa
Application of Neural Networks to Characterization
of Chemical Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Mahmoud Zaki Iskandarani
Application of Machine Learning in Deception Detection . . . . . . . . . . . . 61
Owolafe Otasowie
A New Approach to Estimate the Discharge Coefficient
in Sharp-Crested Rectangular Side Orifices Using Gene
Expression Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Hossein Bonakdari, Bahram Gharabaghi, Isa Ebtehaj, and Ali Sharifi
DiaTTroD: A Logical Agent Diagnostic Test for Tropical Diseases . . . . 97
Sandra Mae W. Famador and Tardi Tjahjadi
A Weighted Combination Method of Multiple K-Nearest Neighbor
Classifiers for EEG-Based Cognitive Task Classification . . . . . . . . . . . . 116
Abduljalil Mohamed, Amer Mohamed, and Yasir Mustafa

vii
viii Contents

Detection and Localization of Breast Tumor in 2D Using


Microwave Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Abdelfettah Miraoui, Lotfi Merad Sidi, and Mohamed Meriah
Regression Analysis of Brain Biomechanics Under
Uniaxial Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
O. Abuomar, F. Patterson, and R. K. Prabhu
Exudate-Based Classification for Detection of Severity of Diabetic
Macula Edema . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Nandana Prabhu, Deepak Bhoir, Nita Shanbhag, and Uma Rao
Analysis and Detection of Brain Tumor Using U-Net-Based
Deep Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Vibhu Garg, Madhur Bansal, A. Sanjana, and Mayank Dave
Implementation of Deep Neural Networks in Facial Emotion
Perception in Patients Suffering from Depressive Disorder: Promising
Tool in the Diagnostic Process and Treatment Evaluation . . . . . . . . . . . 174
Krzysztof Michalik and Katarzyna Kucharska
Invisibility and Fidelity Vector Map Watermarking Based
on Linear Cellular Automata Transform . . . . . . . . . . . . . . . . . . . . . . . . 185
Saleh Al-Ardhi, Vijey Thayananthan, and Abdullah Basuhail
Implementing Variable Power Transmission Patterns
for Authentication Purposes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
Hosam Alamleh, Ali Abdullah S. Alqahtani, and Dalia Alamleh
SADDLE: Secure Aerial Data Delivery with Lightweight Encryption . . . 204
Anthony Demeri, William Diehl, and Ahmad Salman
Malware Analysis with Machine Learning for Evaluating the Integrity
of Mission Critical Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
Robert Heras and Alexander Perez-Pons
Enhanced Security Using Elasticsearch and Machine Learning . . . . . . . 244
Ovidiu Negoita and Mihai Carabas
Memory Incentive Provenance (MIP) to Secure the Wireless Sensor
Data Stream . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
Mohammad Amanul Islam
Tightly Close It, Robustly Secure It: Key-Based Lightweight Process
for Propping up Encryption Techniques . . . . . . . . . . . . . . . . . . . . . . . . 278
Muhammed Jassem Al-Muhammed, Ahmad Al-Daraiseh,
and Raed Abuzitar
Contents ix

Statistical Analysis to Optimize the Generation of Cryptographic Keys


from Physical Unclonable Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
Bertrand Cambou, Mohammad Mohammadi, Christopher Philabaum,
and Duane Booher
Towards an Intelligent Intrusion Detection System:
A Proposed Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
Raghda Fawzey Hriez, Ali Hadi, and Jalal Omer Atoum
LockChain Technology as One Source of Truth for Cyber,
Information Security and Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
Yuri Bobbert and Nese Ozkanli
Introduction of a Hybrid Monitor for Cyber-Physical Systems . . . . . . . 348
J. Ceasar Aguma, Bruce McMillin, and Amelia Regan
Software Implementation of a SRAM PUF-Based Password Manager . . . 361
Sareh Assiri, Bertrand Cambou, D. Duane Booher,
and Mohammad Mohammadinodoushan
Contactless Palm Vein Authentication Security Technique for Better
Adoption of e-Commerce in Developing Countries . . . . . . . . . . . . . . . . . 380
Sunday Alabi, Martin White, and Natalia Beloff
LightGBM Algorithm for Malware Detection . . . . . . . . . . . . . . . . . . . . 391
Mouhammd Al-kasassbeh, Mohammad A. Abbadi,
and Ahmed M. Al-Bustanji
Exploiting Linearity in White-Box AES with Differential
Computation Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
Jakub Klemsa and Martin Novotný
Immune-Based Network Dynamic Risk Control Strategy Knowledge
Ontology Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
Meng Huang, Tao Li, Hui Zhao, Xiaojie Liu, and Zhan Gao
Windows 10 Hibernation File Forensics . . . . . . . . . . . . . . . . . . . . . . . . . 431
Ahmad Ghafarian and Deniz Keskin
Behavior and Biometrics Based Masquerade Detection
Mobile Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
Pranieth Chandrasekara, Hasini Abeywardana, Sammani Rajapaksha,
Sanjeevan Parameshwaran, and Kavinga Yapa Abeywardana
Spoofed/Unintentional Fingerprint Detection Using Behavioral
Biometric Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
Ammar S. Salman and Odai S. Salman
Enabling Paratransit and TNC Services with Blockchain Based
Smart Contracts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
Amari N. Lewis and Amelia C. Regan
x Contents

A Review of Cyber Security Issues in Hospitality Industry . . . . . . . . . . 482


Neda Shabani and Arslan Munir
Extended Protocol Using Keyless Encryption Based on Memristors . . . . 494
Yuxuan Zhu, Bertrand Cambou, David Hely, and Sareh Assiri
Recommendations for Effective Security Assurance
of Software-Dependent Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511
Jason Jaskolka
On Generating Cancelable Biometric Templates Using Visual
Secret Sharing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532
Manisha and Nitin Kumar
An Integrated Safe and Secure Approach for Authentication and
Secret Key Establishment in Automotive Cyber-Physical Systems . . . . . 545
Naresh Kumar Giri, Arslan Munir, and Joonho Kong
How Many Clusters? An Entropic Approach to Hierarchical
Cluster Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 560
Sergei Koltcov, Vera Ignatenko, and Sergei Pashakhin
Analysis of Structural Liveness and Boundedness in Weighted
Free-Choice Net Based on Circuit Flow Values . . . . . . . . . . . . . . . . . . . 570
Yojiro Harie and Katsumi Wasaki
Classification of a Pedestrian’s Behaviour Using Dual Deep
Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581
James Spooner, Madeline Cheah, Vasile Palade, Stratis Kanarachos,
and Alireza Daneshkhah
Towards Porting Astrophysics Visual Analytics Services
in the European Open Science Cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . 598
Eva Sciacca, Fabio Vitello, Ugo Becciani, Cristobal Bordiu,
Filomena Bufano, Antonio Calanducci, Alessandro Costa, Mario Raciti,
and Simone Riggi
Computer Graphics-Based Analysis of Anterior Cruciate
Ligament in a Partially Replaced Knee . . . . . . . . . . . . . . . . . . . . . . . . . 607
Ahmed Imran
An Assessment Algorithm for Evaluating Students Satisfaction
in e-Learning Environments: A Case Study . . . . . . . . . . . . . . . . . . . . . . 613
M. Caramihai, Irina Severin, and Ana Maria Bogatu
The Use of New Technologies in the Organization
of the Educational Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 622
Y. A. Daineko, N. T. Duzbayev, K. B. Kozhaly, M. T. Ipalakova,
Zh. M. Bekaulova, N. Zh. Nalgozhina, and R. N. Sharshova
Contents xi

Design and Implementation of Cryptocurrency Price


Prediction System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 628
Milena Karova, Ivaylo Penev, and Daniel Marinov
Strategic Behavior Discovery of Multi-agent Systems Based
on Deep Learning Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 644
Boris Morose, Sabina Aledort, and Gal Zaidman
Development of Prediction Methods for Taxi Order Service
on the Basis of Intellectual Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . 652
N. A. Andriyanov
Discourse Analysis on Learning Theories and AI . . . . . . . . . . . . . . . . . . 665
Rosemary Papa, Karen Moran Jackson, Ric Brown, and David Jackson
False Asymptotic Instability Behavior at Iterated Functions
with Lyapunov Stability in Nonlinear Time Series . . . . . . . . . . . . . . . . . 673
Charles Roberto Telles
The Influence of Methodological Tools on the Diagnosed Level
of Intellectual Competence in Older Adolescents . . . . . . . . . . . . . . . . . . 694
Sipovskaya Yana Ivanovna
The Automated Solar Activity Prediction System (ASAP) Update
Based on Optimization of a Machine Learning Approach . . . . . . . . . . . 702
Ali K. Abed and Rami Qahwaji

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 719


Preventing Neural Network Weight
Stealing via Network Obfuscation

Kálmán Szentannai, Jalal Al-Afandi, and András Horváth(B)

Faculty of Information Technology and Bionics, Peter Pazmany Catholic University,


Práter u. 50/A, Budapest 1083, Hungary
horvath.andras@itk.ppke.hu

Abstract. Deep Neural Networks are robust to minor perturbations of


the learned network parameters and their minor modifications do not
change the overall network response significantly. This allows space for
model stealing, where a malevolent attacker can steal an already trained
network, modify the weights and claim the new network his own intel-
lectual property. In certain cases this can prevent the free distribution
and application of networks in the embedded domain. In this paper, we
propose a method for creating an equivalent version of an already trained
fully connected deep neural network that can prevent network stealing,
namely, it produces the same responses and classification accuracy, but
it is extremely sensitive to weight changes.

Keywords: Neural networks · Networks stealing · Weight stealing ·


Obfuscation

1 Introduction
Deep neural networks are employed in an emerging number of tasks, many of
which were not solvable before with traditional machine learning approaches. In
these structures, expert knowledge which is represented in annotated datasets is
transformed into learned network parameters known as network weights during
training.
Methods, approaches and network architectures are distributed openly in
this community, but most companies protect their data and trained networks
obtained from tremendous amount of working hours annotating datasets and
fine-tuning training parameters.
Model stealing and detection of unauthorized use via stolen weights is a key
challenge of the field as there are techniques (scaling, noising, fine-tuning, distil-
lation) to modify the weights to hide the abuse, while preserving the functionality
and accuracy of the original network. Since networks are trained by stochastic
optimization methods and are initialized with random weights, training on a
dataset might result various different networks with similar accuracy.
There are several existing methods to measure distances between network
weights after these modifications and independent trainings: [1–3] Obfuscation of
c Springer Nature Switzerland AG 2020
K. Arai et al. (Eds.): SAI 2020, AISC 1230, pp. 1–11, 2020.
https://doi.org/10.1007/978-3-030-52243-8_1
2 K. Szentannai et al.

neural networks was introduced in [4], which showed the viability and importance
of these approaches. In this paper the authors present a method to obfuscate
the architecture, but not the learned network functionality. We would argue that
most ownership concerns are not raised because of network architectures, since
most industrial applications use previously published structures, but because of
network functionality and the learned weights of the network.
Other approaches try to embed additional, hidden information in the network
such as hidden functionalities or non-plausible, predefined answers for previously
selected images (usually referred as watermarks) [5,6]. In case of a stolen network
one can claim ownership of the network by unraveling the hidden functionality,
which can not just be formed randomly in the structure. A good summary com-
paring different watermarking methods and their possible evasions can be found
in [7].
Instead of creating evidence, based on which relation between the original
and the stolen, modified model could be proven, we have developed a method
which generates a completely sensitive and fragile network, which can be freely
shared, since even minor modification of the network weights would drastically
alter the networks response.
In this paper, we present a method which can transform a previously trained
network into a fragile one, by extending the number of neurons in the selected
layers, without changing the response of the network. These transformations can
be applied in an iterative manner on any layer of the network, except the first and
the last layers (since their size is determined by the problem representation). In
Sect. 2 we will first introduce our method and the possible modifications on stolen
networks and in Sect. 3 we will describe our simulations and results. Finally in
Sect. 4 we will conclude our results and describe our planned future work.

2 Mathematical Model of Unrobust Networks


2.1 Fully Connected Layers
In this section we would like to present our method, how a robust network
can be transformed into a non-robust one. We have chosen fully connected net-
works because of their generality and compact mathematical representation.
Fully connected networks are generally applied covering the whole spectrum of
machine learning problems from regression through data generation to classi-
fication problems. The authors can not deny the fact, that in most practical
problems convolutional networks are used, but we would like to emphasize the
following properties of fully connected networks: (1) in those cases when there
is no topographic correlation in the data, fully connected networks are applied
(2) most problems also contain additional fully connected layers after the fea-
ture extraction of the convolutional or residual layers (3) every convolutional
network can be considered as a special case of fully connected ones, where all
weights outside the convolutional kernels are set to zero.
A fully connected deep neural network might consist of several hidden lay-
ers each containing certain number of neurons. Since all layers have the same
Preventing Neural NetworkWeight Stealing via Network Obfuscation 3

architecture, without the loss of generality, we will focus here only on three con-
secutive layers in the network (i − 1, i and i + 1). We will show how neurons
in layer i can be changed, increasing the number of neurons in this layer and
making the network fragile, meanwhile keeping the functionality of the three
layers intact. We have to emphasize that this method can be applied on any
three layers, including the first and last three layers of the network and also that
it can be applied repeatedly on each layer, still without changing the overall
functionality of the network.
The input of the layer i, the activations of the previous layer (i − 1) can be
noted by the vector xi−1 containing N elements. The weights of the network
are noted by the weight matrix Wi and the bias bi where W is a matrix of
N × K elements, creating a mapping RN → RK and bi is a vector containing K
elements. The output of layer i, also the input of layer i + 1 can be written as:

xi = φ(WiN ×K xi−1 + bi ) (1)

where φ is the activation function of the neurons.


The activations of layer i + 1 can be extended as using Eq. 1:

xi+1 = φ(φ(xWi−1N ×K + bi−1 )WiK×L + bi ) (2)

Creating a mapping RN → RL .
One way of identifying a fully connected neural network is to represent it as a
sequence of synaptic weights. Our assumption was that in case of model stealing
certain application of additive noise on the weights would prevent others to reveal
the attacker and conceal thievery. Since fully connected networks are known to be
robust against such modifications, the attacker could use the modified network
with approximately the same classification accuracy. Thus, our goal was to find
a transformation that preserves the loss and accuracy rates of the network, but
introduces a significant decrease in terms of the robustness against parameter
tuning. In case of a three-layered structure one has to preserve the mapping
between the first and third layers (Eq. 2) to keep the functionality of this three
consecutive layers, but the mapping in Eq. 1 (the mapping between the first and
second, or second and third layers), can be changed freely.
Also, our model must rely on an identification mechanism based on a repre-
sentation of the synaptic weights. Therefore, the owner of a network should be
able to verify the ownership based on the representation of the neural network,
examining the average distance between the weights [7].

2.2 Decomposing Neurons



We would like to find such Wi−1 N ×M
and WiM ×L (M ∈ N, M > K) matrices, for
which:
φ(φ(xWi−1N ×K + bi−1 )WiK×L + bi )
 (3)
= φ(φ(xWi−1 N ×M
+ bi−1 )WiM ×L + bi )
4 K. Szentannai et al.

Considering the linear case when φ(x) = x we obtain the following form:
xWi−1N ×K WiK×L + bi−1 WiK×L + bi
 (4)
= xWi−1 N ×M
WiM ×L + bi−1 WiM ×L + bi

The equation above holds only for the special case of φ(x) = x, however in
most cases nonlinear activation functions are used. We have selected the rectified
linear unit (ReLU) for our investigation (φ(x) = max(0, x)). This non-linearity
consist of two linear parts, which means that a variable could be in a linear
domain of Eq. 3 resulting selected lines of 4 (if x ≥ 0), or the equation system
is independent from the variable if the activation function results a constant
zero (if x ≤ 0). This way ReLU gives a selection of given variables (lines) of 4.
However, applying the ReLU activation function has certain constraints.
Assume, that a neuron with the ReLU activation function should be replaced
by two other neurons. This can be achieved by using an α ∈ (0, 1) multiplier:
n
l
φ( Wji xi + blj ) = Njl (5)
i=1

Njl = αNjl + (1 − α)Njl (6)


where αNjl and (1 − α)Njl correspond to the activation of the two neurons.
For each of these, the activation would only be positive if the original neuron
had a positive activation, otherwise it would be zero, this means that all the
decomposed neuron must have the same bias.
After decomposing a neuron, it is needed to choose the appropriate weights
on the subsequent layer. A trivial solution is to keep the original synaptic weights
l+1
represented by the W j column vector. This would lead to the same activation
since
l+1 l+1 l+1
Njl W j = αNjl W j + (1 − α)Njl W j (7)
A fragile network can be created by choosing the same synaptic weights for
the selected two neurons, but it would be easy to spot by the attacker, thus
another solution is needed. In order to find a nontrivial solution we constructed
a linear equation system that can be described by equation system Ap = c,
where A contains the original, already decomposed synaptic weights of the first
layer, meanwhile, p represents the unknown synaptic weights of the subsequent
layer. Vector c contains the corresponding weights from the original network
multiplied together: each element represents the amount of activation related
to one input. Finally the non-trivial solution can be obtained by solving the
following non-homogeneous linear equation system for each output neuron where
index j denotes the output neuron.
⎡  
1
⎤ ⎡ k ⎤
1
w11 w211
... wm1 ⎡ 2 ⎤ wji2 1
wi1
⎢ w 1 1 1 ⎥ wj1 ⎢ i=1
k 2 1 ⎥
⎢ 12 w22 ... wm2 ⎥ ⎢ 2 ⎥ ⎢ i=1 wji wi2 ⎥
⎢ . .. ⎥ ⎢ wj2 ⎥ ⎢ ⎥
⎢ . .. .. ⎥ ⎢ ⎥ ⎢ .. ⎥
⎢ . . . . ⎥ × ⎢ .. ⎥ = ⎢ . ⎥ (8)
⎢ 1 1  ⎥ ⎣ . ⎦ ⎢  k 1 ⎥
⎣w1n w2n ... 1 ⎦
wmn  ⎣ w 2
ji wik ⎦
  
2
wjm i=1
k 2 1
b11 b12 ... b1m i=1 wji bi
Preventing Neural NetworkWeight Stealing via Network Obfuscation 5

It is important to note, that all the predefined weights on layer l + 1 might


change. In summary, this step can be considered as the replacement of a layer,
changing all synaptic weights connecting from and to this layer, but keeping the
biases of the neurons and the functionality of the network intact.
The only constraint of this method is related to the number of neurons regard-
ing the two consecutive layers. It is known, that for matrix A with the size of
M ×N , equation Ap = c has a solution if and only if rank(A) = rank[A|c] where
[A|c] is the extended matrix. The decomposition of a neuron described in Eq. 7
results in linearly dependent weight vectors on layer l, therefore when solving the
equation system the rank of the matrix A is less than or equal to min(N + 1, K).
If the rank is equal to N + 1 (meaning that K ≥ N + 1) then vector c with the
dimension of N +1 would not introduce a new dimension to the subspace defined
by matrix A. However if rank(A) = K (meaning that K ≤ N + 1) then vector
c could extend the subspace defined by A. Therefore, the general condition for
solving the equation system is: K ≥ N + 1.
This shows that one could increase the number of the neurons in a layer,
and divide the weights of the existing neuron in that layer. We have used this
derivation and aim to find a solution of Eq. 7 where the order of magnitudes are
significantly different (in the range of 106 ) for both the network parameters and
for the eigenvalues of the mapping RN → RL .

2.3 Introducing Deceptive Neurons


The method described in the previous section results a fragile neural network, but
unfortunately it is not enough to protect the network weights, since an attacker
could identify the decomposed neurons based on their biases or could fit a new
neural network on the functionality implemented by the layer. To prevent this
we will introduce deceptive neurons in layers. The purpose of these neurons is to
have non-zero activation in sum if and only if noise was added to their weights
apart from this all these neurons have to cancel each others effect out in the
network, but not necessarily in a single layer.
The simplest method is to define a neuron with an arbitrary weight and a bias
of an existing neuron resulting a large activation and making a copy of it with
the difference of multiplying the output weights by −1. As a result, these neurons
do not contribute to the functionality of the network. However, adding noise to
the weights of these neurons would have unexpected consequences depending on
the characteristics of the noise, eventually leading to a decrease of classification
accuracy.
One important aspect of this method is to hide the generated neurons and
obfuscate the network to prevent the attacker to easily filter our deceptive neu-
rons in the architecture. Choosing the same weights again on both layers would
be an obvious sign to an attacker, therefore this method should be combined
with decomposition described in Sect. 2.2.
Since decomposition allows the generation of arbitrarily small weights one
can select a suitably small magnitude, which allows the generation of R real
(non deceptive) neurons in the system, and half of their weights (α parameters)
6 K. Szentannai et al.

can be set arbitrarily, meanwhile the other half of the weights will be determined
by Eq. 8. For each real neuron one can generate a number (F ) of fake neurons
forming groups of R number of real and F number of fake neurons. These groups
can be easily identified in the network since all of them will have the same bias,
but the identification of fake and real neurons in a group is non-polynomial.
The efficiency of this method should be measured in the computational com-
plexity of successfully finding two or more corresponding fake neurons having
a total activation of zero in a group. Assuming Lthat
 only  one pair of fake neu-
rons was added to the network, it requires i=0 Ri +F 2
i
steps to successfully
identify the fake neurons, where Ri + Fi denotes the number of neurons in the
corresponding hidden layer, and L is the number of hidden layers. This can be
further increased by decomposing L thei +F
fake neurons using Eq. 8: in that case the
required number of steps is i=0 Rd+2 i
, d being the number of extra decom-
posed neurons. This can be maximized if d + 2 = Ri + Fi /2, where i denotes
the layer, where the fake neurons are located. However, this is true only if the
attacker has information about the number of deceptive neurons. Without any
prior knowledge, the attacker has to guess the number of deceptive neurons
as well (0, 1, 2 . . . Ri + Fi − 1) which leads to exponentially increasing computa-
tional time.

3 Experiments
3.1 Simulation of a Simple Network
As a case study we have created a simple fully connected neural network with
three layers, each containing two neurons to present the validity of our approach.
The functionality of the network can be considered as a mapping f : R2 → R2 .

   
6 −1  5 3 
w1 = , b1 = 1 −5 w2 = , b2 = 7 1
−1 7 9 −1
We added two neurons to the hidden layer with decomposition, which does
not modify the input and output space and no deceptive neurons were used in
this experiment. After applying the methods described in Sect. 2.1, we obtained
a solution of:

 
0.0525 −0.4213 6.0058 −0.5744
w1 =
−0.0087 2.9688 −0.9991 4.0263

b1 = 0.0087 −2.1066 1.0009 −2.8722
⎡ ⎤
4.1924e + 03 −5.4065e + 03
⎢ −2.3914 7.3381 ⎥
w2 = ⎢
⎣ −3.2266


5.7622
6.9634 −7.0666

b2 = 7 1
Preventing Neural NetworkWeight Stealing via Network Obfuscation 7

Fig. 1. This figure depicts the response of a simple two-layered fully connected network
for a selected input (red dot) and the response of its variants with %1 noise (yellow
dots) added proportionally to the weights. The blue dots represent the responses of the
transformed MimosaNets under the same level of noise on their weights, meanwhile the
response of the transformed network (without noise) remained exactly the same.

In the following experiment we have chosen an arbitrary input vector: [7,9].


We have measured the response of the network for this input, each time intro-
ducing 1% noise to the weights of the network. Figure 1 shows the response of the
original network and the modified network after adding 1% noise. The variances
of the original network for the first output dimension is 6.083 and 8.399 for the
second, meanwhile the variances are 476.221 and 767.877 for the decomposed
networks respectively. This example demonstrates how decomposition of a layer
can increase the networks dependence on its weights.

3.2 Simulations on the MNIST Dataset

We have created a five layered fully connected network containing 32 neurons in


each hidden-layer (and 728 and 10 neurons in the input and output layers) and
trained it on the MNIST [8] dataset, using batches of 32 and Adam Optimizer
[9] for 7500 iterations. The network has reached an accuracy of 98.4% on the
independent test set.
We have created different modifications of the network by adding 9,18,36,72
extra neurons. These neurons were divided equally between the three hidden-
layers and 2/3 of them were deceptive neurons (since they were always created
in pairs) and 1/3 of them were created by decomposition. This means that in
case of 36 additional neurons 2 × 4 deceptive neurons were added to each layer
and four new neurons per layer were created by decomposition.
8 K. Szentannai et al.

In our hypothetical situations these networks (along with the original) could
be stolen by a malevolent attacker, who would try to conceal his thievery by
using the following three methods: adding additive noise proportionally to the
network weights, continuing network training on arbitrary data and network
knowledge distillation. All reported datapoints are an average of 25 independent
measurements.

Dependence on Additive Noise. We have investigated network performance


using additive noise to the network weights. The decrease of accuracy which
depends on the ratio of the additive noise can be seen in Fig. 2.
At first we have tested a fully connected neural network trained on the
MNIST dataset without making modifications to it. The decrease of accuracy
was not more than 0.2% even with a relatively high 5% noise. This shows the
robustness of a fully connected network.
After applying the methods described in Sect. 2 network accuracy retro-
gressed to 10% even in case of noise which was less than 1% of the network
weights, as Fig. 2 depicts. This alone would reason the applicability of our
method, but we have investigated low level noises further, which can be seen
on Fig. 3. As it can be seen from the figure, accuracy starts to drop when the
ratio of additive noise reaches the level of 10−7 , which means the attacker can
not significantly modify the weights. This effect could be increased by adding
more and more neurons to the network.

Fig. 2. This figure depicts accuracy changes on the MNIST dataset under various level
of additive noise applied on the weights. The original network (purple) is not dependent
on these weight changes, meanwhile accuracies retrogress in the transformed networks,
even with the lowest level of noise.

Dependence on Further Training Steps. Additive noise randomly modi-


fies the weights, but it is important to examine how accuracy changes in case
of structured changes exploiting the gradients of the network. Figure 4 depicts
accuracy changes and average in weights distances by applying further training
Preventing Neural NetworkWeight Stealing via Network Obfuscation 9

steps in the network. Further training was examined using different step sizes
and optimizers (SGD,AdaGrad and ADAM) training the network with original
MNIST and randomly selected labels and the results were qualitatively the same
in all cases.

Dependence on Network Distillation. We have tried to distill the knowledge


in the network and train a new neural network to approximate the functionality
of previously selected layers, by applying the method described in [10].
We have generated one million random input samples with their outputs for
the modified networks and have used this dataset to approximate the function-
ality of the network.

Fig. 3. A logarithmic plot depicting the same accuracy dependence as on Fig. 2, focus-
ing on low noise levels. As it can be seen from the plot, accuracy values do not change
significantly under 10−7 percent of noise, which means the most important values of
the weights would remain intact to proof connection between the original and modified
networks.

We have created three-layered neural networks containing 32, 48, 64, 128
neurons in the hidden layer (The number of neurons in the first and last layer
were determined by the original network) and tried to approximate the function-
ality of the hidden layers of the original structure. Since deceptive neurons have
activations in the same order of magnitude as the original responses, these values
disturb the manifold of the embedded representations learned by the network
and it is more difficult to be approximated by a neural network. Table 1 contains
the maximum accuracies which could be reached with knowledge distillation,
depending on the number of deceptive neurons and the neurons in the architec-
ture used for distillation. This demonstrates, that our method is also resilient
towards knowledge distillation.
10 K. Szentannai et al.

Fig. 4. The figure plots accuracy dependence of the networks in case of further training
(applying further optimization steps). As it can be seen from the plot weights had to
be kept in 10−7 average distance to keep the same level of accuracy.

Table 1. The table displays the maximum accuracies reached with knowledge distil-
lation. The different rows display the number of extra neurons which were added to
the investigated layer, and the different columns show the number of neurons in the
hidden layer of the fully connected architecture, which was used for distillation.

#Deceptive N. #N. = 32 #N. = 48 #N. = 64 #N. = 128


9 0.64 0.65 0.69 0.71
18 0.12 0.14 0.15 0.17
36 0.10 0.11 0.10 0.13
72 0.11 0.09 0.10 0.10

4 Conclusion

In this paper, we have shown a transformation method which can significantly


increase a network’s dependence on its weights, keeping the original functionality
intact. We have also presented how deceptive neurons can be added to a network,
without disturbing its original response. Using these transformations iteratively
one can create and openly share a trained network, where it is computationally
extensive to reverse engineer the original network architecture and embeddings
in the hidden layers. The drawback of the method is the additional computa-
tional need for the extra neurons, but this is not significant, since computational
increase is polynomial.
Preventing Neural NetworkWeight Stealing via Network Obfuscation 11

We have tested our method on simple toy problems and on the MNIST
dataset using fully-connected neural networks and demonstrated that our app-
roach results non-robust networks for the following perturbations: additive noise,
application of further training steps and knowledge distillation.

Acknowledgments. This research has been partially supported by the Hungarian


Government by the following grant: 2018-1.2.1-NKP-00008: Exploring the Mathemat-
ical Foundations of Artificial Intelligence also the funds of grant EFOP-3.6.2-16-2017-
00013 are gratefully acknowledged.

References
1. Koch, E., Zhao, J.: Towards robust and hidden image copyright labeling. In: IEEE
Workshop on Nonlinear Signal and Image Processing, vol. 1174, pp. 185–206,
Greece, Neos Marmaras (1995)
2. Wolfgang, R.B., Delp, E.J.: A watermark for digital images. In: Proceedings of the
International Conference on Image Processing, vol. 3, pp. 219–222. IEEE (1996)
3. Zarrabi, H., Hajabdollahi, M., Soroushmehr, S., Karimi, N., Samavi, S., Najarian,
K.: Reversible image watermarking for health informatics systems using distortion
compensation in wavelet domain (2018 ) arXiv preprintarXiv:1802.07786
4. Xu, H., Su, Y., Zhao, Z., Zhou, Y., Lyu, M.R., King, I.: Deepobfuscation: securing
the structure of convolutional neural networks via knowledge distillation (2018)
arXiv preprint arXiv:1806.10313
5. Namba, R., Sakuma, J.: Robust watermarking of neural network with exponential
weighting (2019) arXiv preprint arXiv:1901.06151
6. Gomez, L., Ibarrondo, A., Márquez, J., Duverger, P.: Intellectual property protec-
tion for distributed neural networks (2018)
7. Hitaj, D., Mancini, L.V.: Have you stolen my model? evasion attacks against deep
neural network watermarking techniques (2018) arXiv preprint arXiv:1809.00615
8. LeCun, Y., Cortes, C., Burges, C.: MNIST handwritten digit database. AT&T
Labs 2 (2010). http://yann.lecun.com/exdb/mnist
9. Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization (2014) arXiv
preprint arXiv:1412.6980
10. Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network
(2015) arXiv preprint arXiv:1503.02531
Applications of Z-Numbers and Neural
Networks in Engineering

Raheleh Jafari1(B) , Sina Razvarz2 , and Alexander Gegov3


1
School of design, University of Leeds, Leeds LS2 9JT, UK
r.jafari@leeds.ac.uk
2
Departamento de Control Automático, CINVESTAV-IPN (National Polytechnic
Institute), Mexico City, Mexico
srazvarz@yahoo.com
3
School of Computing, University of Portsmouth, Buckingham Building, PO1 3HE
Portsmouth, UK
alexander.gegov@port.ac.uk

Abstract. In the real world, much of the information on which deci-


sions are based is vague, imprecise and incomplete. Artificial intelligence
techniques can deal with extensive uncertainties. Currently, various types
of artificial intelligence technologies, like fuzzy logic and artificial neural
network are broadly utilized in the engineering field. In this paper, the
combined Z-number and neural network techniques are studied. Further-
more, the applications of Z-numbers and neural networks in engineering
are introduced.

Keywords: Artificial intelligence · Fuzzy logic · Z-number · Neural


network

1 Introduction
Intelligent systems are composed of fuzzy systems and neural networks. They
have particular properties such as the capability of learning, modeling and resolv-
ing optimizing problems, suitable for specific kind of applications. The intelligent
system can be named hybrid system in case that it combines a minimum of two
intelligent systems. For example, the mixture of the fuzzy system and neural
network causes the hybrid system to be called a neuron-fuzzy system.
Neural networks are made of interrelated groups of artificial neurons that
have information which is obtainable by computations linked to them. Mostly,
neural networks can adapt themselves to structural alterations while the training
phase. Neural networks have been utilized in modeling complicated connections
among inputs and outputs or acquiring patterns for the data [1–12].
Fuzzy logic systems are broadly utilized to model the systems characterizing
vague and unreliable information [13–29]. During the years, investigators have
proposed extensions to the theory of fuzzy logic. Remarkable extension includes
Z-numbers [30]. The Z-number is defined as an ordered pair of fuzzy numbers
c Springer Nature Switzerland AG 2020
K. Arai et al. (Eds.): SAI 2020, AISC 1230, pp. 12–25, 2020.
https://doi.org/10.1007/978-3-030-52243-8_2
Applications of Z-Numbers and Neural Networks in Engineering 13

(C, D), such that C is a value of some variables and D is the reliability which
is a value of probability rate of C. Z-numbers are widely applied in various
implementations in different areas [31–36].
In this paper, the basic principles and explanations of Z-numbers and neu-
ral networks are given. The applications of Z-numbers and neural networks in
engineering are introduced. Also, the combined Z-number and neural network
techniques are studied. The rest of the paper is organized as follows. The the-
oretical background of Z-numbers and artificial neural networks are detailed in
Sect. 2. Comparison analysis of neural networks and Z-number systems is pre-
sented in Sect. 3. The combined Z-number and neural network techniques are
given in Sect. 4. The conclusion of this work is summarized in Sect. 5.

2 Theoretical Background

In this section, we provide a brief theoretical insight of Z-numbers and artificial


neural networks.

2.1 Z-Numbers

Mathematical Preliminaries. Here some necessary definitions of Z-number


theory are given.
Definition 1. If q is: 1) normal, there exists ω 0 ∈  where q(ω0 ) = 1, 2) convex,
q(υω + (1 − υ)ω) ≥ min{q(ω), q(τ )}, ∀ω, τ ∈ , ∀υ ∈ [0, 1], 3) upper semi-
continuous on , q(ω) ≤ q(ω0 ) + , ∀ω ∈ N (ω0 ), ∀ω0 ∈ , ∀ > 0, N (ω0 ) is a
neighborhood, 4) q + = {ω ∈ , q(ω) > 0} is compact, so q is a fuzzy variable,
q ∈ E :  → [0, 1].
The fuzzy variable q is defined as below
 
q = q, q (1)

such that q is the lower-bound variable and q is the upper-bound variable.


Definition 2. The Z-number is composed of two elements Z = [q(ω), p]. q(ω)
is considered as the restriction on the real-valued uncertain variable ω and p is
considered as a measure of the reliability of q. The Z-number is defined as Z + -
number, when q(ω) is a fuzzy number and p is the probability distribution of ω.
If q(ω), and p, are fuzzy numbers, then the Z-number is defined as Z − -number.
The Z + -number has more information in comparison with the Z − -number.
In this work, we use the definition of Z + -number, i.e., Z = [q, p] , q is a fuzzy
number and p is a probability distribution.
The triangular membership function is defined as
 ω−a
b−a a ≤ ω ≤ b otherwise μ = 0
μq = G (a, b, c) = c−ω (2)
c−b b ≤ ω ≤ c
q
14 R. Jafari et al.

and the trapezoidal membership function is defined as


⎧ ω−a
⎨ b−a a ≤ ω ≤ b
μq = G (a, b, c, d) = d−ω c ≤ ω ≤ d otherwise μq = 0 (3)
⎩ d−c
1 b≤ω≤c

The probability measure of q is defined as



P (q) = μq (ω)p(ω)dω (4)


such that p is the probability density of ω. For discrete Z-numbers we have


n
P (q) = μq (ωj )p(ωj ) (5)
j=1

Definition 3. The α-level of the Z-number Z = (q, p) is stated as below

[Z]α = ([q]α , [p]α ) (6)

such that 0 < α ≤ 1. [p]α is calculated by the Nguyen’s theorem


α
[p]α = p([q]α ) = p([q α , q α ]) = P α , P (7)

such that p([q]α ) = {p(ω)|ω ∈ [q]α }. Hence, [Z]α is defined as


α   α
[Z]α = Z α , Z = qα , P α , qα , P (8)
α
such that P α = q α p(ω α α α α α α
j ), P = q p(ω j ), [ωj ] = (ω j , ω j ).
Let Z1 = (q1 , p1 ) and Z2 = (q2 , p2 ), we have

Z12 = Z1 ∗ Z2 = (q1 ∗ q2 , p1 ∗ p2 ) (9)

where ∗ ∈ {⊕, , }. ⊕, and , indicate sum, subtract and multiply, respec-


tively.
The operations utilized for the fuzzy numbers [q1 ]α = [q11
α α
, q12 ] and [q2 ]α =
α α
[q21 , q22 ] are defined as [37],

[q1 ⊕ q2 ]α = [q1 ]α + [q2 ]α = [q11α


+ q21α α
, q12 α
+ q22 ]
α
[q1 q2 ] =  [q1 ] − [q2 ] = [q11 − q22 , q12 − q21
α α α α α α
]
(10)
min{q α α
q , q α α
q , q α α
q , q α α
q }
[q1 q2 ]α = 11 21 11 22 12 21 12 22
max{q11 α α α α
q21 , q11 α α
q22 , q12 q22 }
α α
q21 , q12

For the discrete probability distributions, the following relation is defined for all
p1 ∗ p2 operations

p 1 ∗ p2 = p1 (ω1,j )p2 (ω2,(n−j) ) = p12 (ω) (11)


ι
Applications of Z-Numbers and Neural Networks in Engineering 15

Fig. 1. Membership functions applied for (a) cereal yield, cereal production, economic
growth, (b) threat rate, and (c) reliability

Background and Related Work. The implementations of Z-numbers based


techniques are bounded because of the shortage of effective approaches for cal-
culation with Z-numbers.
In [38], the capabilities of the Z-numbers in the improvement of the quality
of risk assessment are studied. Prediction equal to (High, Very Sure) is institu-
tionalized in the form of Z-evaluation “y is Z(c, p)”, such that y is considered
as a random variable of threat probability, c and p are taken to be fuzzy sets,
demonstrating soft constraints on a threat probability and a partial reliability,
respectively. The likelihood of risk is illustrated by Z-number as: Probability
= Z1 (High, Very Sure), such that c is indicated through linguistic terms High,
Medium, Low, also, p is indicated through terms Very Sure, Sure, etc. Likewise,
consequence rate is explained as: Consequence measure = Z2 (Low, Sure). Threat
rates (Z12 ) is computed as the product of the probability (Z1 ) and consequence
measure (Z2 ).
In [39], Z-number-based fuzzy system is suggested to determine the food
security risk level. The proposed system is relying on fuzzy If-Then rules, which
applies the basic parameters such as cereal production, cereal yield, and economic
growth to specify the threat rate of food security. The membership functions
applied to explain input, as well as output variables, are demonstrated in Fig. 1.
In [40], the application of the Z-number theory to selection of optimal
alloy is illustrated. Three alloys named Ti12Mo2Sn alloy, Ti12Mo4Sn alloy, and
Ti12Mo6Sn alloy are examined and an optimal titanium alloy is selected using
the proposed approach. The optimality of the alloys is studied based on three
criteria: strength level, plastic deformation degree, and tensile strength.
16 R. Jafari et al.

Fig. 2. The structure of a biological neuron

2.2 Neural Networks

Neural networks are constructed from neurons and synapses. They alter their
rates in reply from nearby neurons as well as synapses. Neural networks operate
similar to computer as they map inputs to outputs. Neurons, as well as synapses,
are silicon members, which mimic their treatment. A neuron gathers the total
incoming signals from other neurons, afterward simulate its reply represented
by a number. Signals move among the synapses, which contain numerical rates.
Neural networks learn once they vary the value of their synapsis. The structure
of a biological neuron or nerve cell is shown in Fig. 2. The processing steps inside
each neuron is demonstrated in Fig. 3.

Background and Related Work. In [41], artificial neural network technique


is utilized for modeling the void fraction in two-phase flow inside helical vertical
coils with water as work fluid. In [42] artificial neural network and multi-objective
genetic algorithm are applied for optimizing the subcooled flow boiling in a
vertical pipe. Pressure, the mass flux of the water, inlet subcooled temperature,
as well as heat flux are considered as inlet parameters. The artificial neural
network utilizes inlet parameters for predicting the objective functions, which
are the maximum wall surface temperature as well as averaged vapor volume
fraction at the outlet. The optimization procedure of design parameters is shown
in Fig. 4.
In [43], artificial neural network technique is applied for predicting heat trans-
fer in supercritical water. The artificial neural network is trained on the basis
of 5280 data points gathered from experimental results. Mass flux, heat flux,
pressure, tube diameter, as well as bulk specific enthalpy are taken to be the
inputs of the proposed artificial neural network. The tube wall temperature is
taken to be the output, see Fig. 5.
Applications of Z-Numbers and Neural Networks in Engineering 17

Fig. 3. Processing steps inside each neuron

3 Comparison Analysis of Neural Networks


and Z-Number Systems
Neural networks and Z-number systems can be considered as a part of the soft
computing field. The comparison of Neural networks and Z-number systems is
represented in Table 1. Neural networks have the following advantageous:

Table 1. The comparison of neural networks and Z-number systems.

Z-number systems Neural networks


Knowledge presentation Very good Very bad
Uncertainty tolerance Very good Very good
Inaccuracy tolerance Very good Very good
Compatibility Bad Very good
Learning capability Very bad Very good
Interpretation capability Very good Very bad
Knowledge detection and data mining Bad Very good
Maintainability Good Very good

i Adaptive Learning: capability in learning tasks on the basis of the data sup-
plied to train or initial experience.
ii Self-organization: neural networks are able to create their organization while
time learning.
iii Real-time execution: the calculations of neural networks may be executed in
parallel, also specific hardware devices are constructed, which can capture
the benefit of this feature.

Neural networks have the following drawbacks:


Another Random Document on
Scribd Without Any Related Topics
The Project Gutenberg eBook of L'oeuvre du
divin Arétin, première partie
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: L'oeuvre du divin Arétin, première partie

Author: Pietro Aretino

Editor: Guillaume Apollinaire

Release date: September 27, 2013 [eBook #43823]


Most recently updated: October 23, 2024

Language: French

Credits: Produced by Laurent Vogel, Jean-Adrien Brothier and the


Online Distributed Proofreading Team at
http://www.pgdp.net
(This file was produced from images generously made
available by the Bibliothèque nationale de France
(BnF/Gallica) at http://gallica.bnf.fr)

*** START OF THE PROJECT GUTENBERG EBOOK L'OEUVRE DU


DIVIN ARÉTIN, PREMIÈRE PARTIE ***
L'ŒUVRE DU DIVIN ARÉTIN

==Il a été tiré de cet ouvrage==


10 exemplaires sur Japon Impérial
==========(1 à
10)=========
===25 exemplaires sur
Hollande===
==========(11 à
35)========
100 exemplaires sur papier
d'Arches
==========36 à
135=========

Droits de reproduction
réservés pour tous pays, y
compris la Suède, la
Norvège et le Danemark.
PETRVS ARRETINVS ACERRIMVS VIRTVTVM AC VITIORVM
DEMOSTRATOR
NON MANVS ARTIFICIS MAGE DIGNVM OS PINGERE NON
OS
HOC PINGI POTERAT NOBILIORE MANV
PELLÆVS IVVENIS SI VIVERET HAC VOLO DESTRA
PINGIER HOC TANTVM DICERET ORE CANI

LES MAITRES DE L'AMOUR

L'Œuvre
DU
DIVIN ARÉTIN

Première Partie

Les Ragionamenti

La Vie des Nonnes;—La Vie des Femmes mariées;


La Vie des Courtisanes;

Sonnets Luxurieux

TRADUCTIONS NOUVELLES ET MORCEAUX TRADUITS


POUR LA PREMIÈRE FOIS

INTRODUCTION ET NOTES

PAR

Guillaume APOLLINAIRE
PARIS

BIBLIOTHÈQUE DES CURIEUX

4, RUE DE FURSTENBERG, 4

MCMIX
INTRODUCTION
Un singulier cours d'eau à double pente coule dans le val que
domine Arezzo: c'est la Chiana. Elle peut être donnée comme une
image de ce Pierre dit l'Arétin, qui, à cause de sa gloire et de son
déshonneur, est devenu l'une des figures les plus attachantes du xvie
siècle. Elle est, en même temps, une des plus mal connues. A vrai
dire, si de son vivant même la renommée de l'Arétin n'alla pas sans
infamie, après sa mort on chargea sa mémoire de tous les péchés de
son époque. On ne comprenait pas comment l'auteur des
Ragionamenti pouvait avoir écrit Les Trois Livres de l'Humanité du
Christ, l'on se demandait comment ce débauché avait pu être l'ami
des souverains, des papes et des artistes de son temps. Ce qui
devait le justifier aux yeux de la postérité a été cause de sa
condamnation. En fait de génie, on ne lui a laissé que celui de
l'intrigue. Je m'étonne même qu'on ne l'ait pas accusé d'avoir acquis
ses biens et son crédit par la magie.
Ce Janus bifronts a déconcerté la plupart de ses biographes et de
ses commentateurs. Son nom seul, depuis plus de trois siècles,
effraye les plus bénévoles. Il demeure l'homme des postures, non
pas à cause de ses Sonnets, mais par la faute d'un dialogue en prose
qu'il n'a point écrit et où on en indique 35. Cependant, le populaire
n'en met que 32 sur le compte de l'imagination luxurieuse du Divin.
En Italie, les lettrés le voient d'un mauvais œil; les érudits
n'abordent des recherches sur cet homme qu'avec beaucoup de
répugnance et ne prononcent son nom que du bout des lèvres,
osant à peine feuilleter ses livres du bout des doigts. Chez nous, les
gens du monde accouplent sa mémoire à celle du marquis de Sade;
les collégiens, à celle d'Alfred de Musset; pour le peuple et la petite
bourgeoisie, son nom évoque encore, avec ceux de Boccace et de
Béranger, la grivoiserie qui est toute la santé et la sauvegarde du
mariage. C'est que la variété est bien la seule arme que l'on possède
contre la satiété. Et l'homme qui, directement ou indirectement, a
fourni à l'amour un prétexte pour ne point lasser devrait être honoré
par tous les amants et surtout par les gens mariés. Sans doute, on
connaîtrait les postures, même si le dialogue attribué à l'Arétin
n'avait pas été écrit, mais on n'en connaîtrait pas autant, et ni
Forberg, ni les livres hindous, ni les autres manuels d'érotologie qui
en indiquent un nombre beaucoup plus considérable ne seront
jamais assez populaires pour donner à l'époux et à l'épouse une
occasion naturelle, provenant d'une locution quasi proverbiale, de
repousser l'ennui en variant les plaisirs. L'Arétin, qui utilisa le premier
cette arme moderne, la Presse, qui, le premier, sut modifier l'opinion
publique, qui exerça une influence sur le génie de Rabelais et peut-
être sur celui de Molière[1], est aussi, par aventure, le maître de
l'Amour occidental[2]. Il est devenu une sorte de demi-dieu fescennin
qui a remplacé Priape dans le Panthéon populaire d'aujourd'hui. On
l'invoque ou on l'évoque au moment de l'amour, car pour ce qui
regarde ses ouvrages, on ne les connaît pas. Les exemplaires en
sont devenus rares. En Italie même, on ne connaît guère que son
théâtre. Les Ragionamenti n'avaient jamais été traduits en français
avant que Liseux en publiât le texte accompagné de la traduction
d'Alcide Bonneau[3] d'après laquelle fut faite la tradition anglaise
publiée par le même éditeur. Elle dut servir de modèle au Dr Heinrich
Conrad pour la première et toute récente édition allemande:
Gespräche des Göttlichen Aretino, éditée par l'Insel Verlag de
Leipzig.
Ajoutons qu'une partie de l'œuvre arétinesque est aujourd'hui
perdue; une autre demeure inédite dans les recueils manuscrits
dispersés dans les Bibliothèques européennes; une autre enfin lui
appartient sans doute aussi qui ne lui est pas attribuée.

Pietro Aretino naquit à Arezzo, en Toscane, pendant la nuit du 19 au


20 avril 1492, nuit du jeudi au vendredi saints, quelques mois avant
la découverte de l'Amérique, et mourut à Venise, le 21 octobre
1556[4].
Avec une singulière précision, le catalogue imprimé de la
Bibliothèque Nationale l'appelle: Pietro Bacci, dit Aretino. Les raisons
qu'on avait alléguées pour soutenir l'opinion abandonnée aujourd'hui
que l'Arétin avait eu pour père un gentilhomme d'Arezzo nommé
Luigi Bacci n'autorisaient nullement les bibliographes de la Nationale
à accorder ce nom à Messer Pietro, qui de toute façon n'aurait été
qu'un bâtard de Bacci, n'ayant jamais porté ce nom. C'est aussi sans
fondement qu'on l'a gratifié de noms comme Della Bura ou De'Burali,
Bonci, Bonamici, Camaiani, etc.
On sait maintenant que le père de l'Arétin était un pauvre cordonnier
d'Arezzo, nommé Luca. Les recherches de M. Alessandro Luzio dans
les archives de Florence ne laissent plus aucun doute à cet égard[5].
En 1550, un certain Medero Nucci, d'Arezzo, vient chercher fortune à
Venise. Et d'abord son compatriote, l'Arétin, le protège, le présente à
l'ambassadeur du duc de Florence. Puis tout se gâte; l'Arétin écrit à
l'ambassadeur de s'en défier, alléguant des désordres et des
scandales dans la vie privée de Medero Nucci, qui pour se venger
envoie à l'Arétin un cartel de défi où il lui reproche d'avoir écrit les
sonnets sous les figures de Raphaello da Orbino, le Trentunno, La
Puttana errante, les Six journées. Et cette missive est adressée Allo
Aretino Pietro de Lucha, calzolaio a Venezia, c'est-à-dire A l'Arétin
Pierre (fils) de Lucha, cordonnier à Venise. Voici donc le nom du père
de notre Pierre: Lucha ou Luca, Luc en français. D'ailleurs le Divin ne
renie pas une origine aussi obscure. Il envoie au duc Côme la lettre
de Nucci et lui en écrit:
«...Pour en venir maintenant à la mention de sa maudite
épistole, je dis que je me glorifie du titre qu'il me donne
pour m'avilir, car il enseigne ainsi aux nobles à procréer
des fils semblables à celui qu'un cordonnier a engendré
dans Arezzo.»
Quel orgueil! ne croirait-on pas entendre un des maréchaux de
Napoléon se glorifier de n'avoir pour aïeux que des gens du peuple?
Ce sont ces lettres qu'a retrouvées M. Alessandro Luzio. Elles ne
nous renseignent d'ailleurs que touchant le prénom et l'état du père
de l'Arétin. Et nous ne sommes pas pour cela plus avancés au sujet
du nom de famille de notre Pierre. Il est fort possible au demeurant
que le cordonnier Lucas n'eût pas d'autre nom. Il se peut également
que ce fût le nom de la famille du Divin. Luca est de nos jours
encore un nom patronymique très répandu non seulement en Italie,
mais encore en Corse. Il ne semble pas, d'autre part, que l'Arétin se
soit jamais ouvert à qui que ce soit touchant le nom de son père et
en ait fait mention. Cependant, je crois être en mesure d'indiquer
dans un giudicio retrouvé et publié par M. Alessandro Luzio[6] un
passage dans lequel en 1534, longtemps avant le message de Nucci,
le Divino mentionnait le nom paternel en équivoquant. Au temps de
l'Arétin, l'astrologie judiciaire était florissante. Au commencement de
chaque année, les astrologues publiaient leurs giudicii ou pronostics.
Avec cette prescience du rôle que devait jouer plus tard la Presse et
à cause de laquelle Philarète Chasles eut raison de voir en lui un
précurseur du journalisme, l'Arétin comprit le parti qu'on pouvait
tirer de ces libelles pour former l'opinion publique. Il écrivit plusieurs
de ces giudicii satiriques et d'ailleurs peu prophétiques, tous perdus
jusqu'à ces dernières années, sauf quelques fragments. A cette
heure, on possède en entier celui qu'a publié récemment M.
Alessandro Luzio et qui provient d'un manuscrit de la fin du xvie
siècle, copié par un Allemand et conservé à Vienne, en Autriche.
Tout laisse croire que le copiste allemand a eu sous les yeux un
imprimé. C'est l'avis de M. Luzio, qui n'est pas d'accord sur ce point
avec les autres arétiniens d'Italie. En effet, on ne connaît aucun
exemplaire imprimé des giudicii de l'Arétin. Et, cependant, les raisons
de M. Luzio me semblent bonnes. Des pamphlets comme celui qui
nous occupe ne pouvaient avoir d'effet sur l'opinion publique (et
c'est à cela qu'ils étaient destinés) que s'ils étaient répandus à un
grand nombre d'exemplaires, et l'on sait que l'Arétin a fait publier à
part plusieurs de ses lettres sur les grands événements de son
temps.
D'autre part, M. Luzio, qui a vu le manuscrit de Vienne, affirme que
le copiste allemand devait connaître mal l'italien et n'aurait pu copier
aussi correctement un manuscrit. Il aurait donc eu entre les mains
un imprimé perdu aujourd'hui. Quoi qu'il en soit, en 1534, l'Arétin
tenait encore pour François Ier dont il attaque, dans son pronostic,
tous les ennemis, à commencer par Charles-Quint, dans le parti
duquel il allait bientôt passer. Aussi dédie-t-il son pamphlet Alla
Sacra Maesta Christianissima et il l'intitule: Pronostico dell'anno
MDXXXIIII composto da Pietro Aretino, Flagelle dei Principi et quinto
evangelista. Ce n'est pas au hasard que l'Arétin se targue de cette
dernière qualité. Pourquoi s'appellerait-il cinquième évangéliste?... Il
y a là-dessous un jeu de mots dont il nous donne la clef au
paragraphe 31: Del Flagello dei Principi, qui commence ainsi: Pierre
Arétin qui eut comme ascendants Luc, Jean, Marc et Mathieu[7]... En
effet, y ayant quatre évangélistes, Pierre Arétin, fils de Luca ou Luc,
l'un d'eux, c'est-à-dire venant après lui, peut bien prétendre être le
cinquième évangéliste, si l'on veut bien entendre par évangéliste un
prophète. L'Arétin n'a pu résister au plaisir d'équivoquer d'une façon
assez embarrassée sur le nom de son père le cordonnier et pour cela
il n'a pas hésité à changer l'ordre des quatre évangélistes et à
torturer le sens de ce mot. Et c'est la seule mention connue, pensé-
je, que l'Arétin ait faite du nom de son père[8].
L'Arétin ne se vantait pas à tout propos de son origine plébéienne.
On lui a reproché de ne pas s'être beaucoup occupé de son père. Et
les sarcasmes du Franco, du Doni et du Berni touchant le métier de
cordonnier qu'exerçait le bonhomme nous montrent assez combien
ces allusions devaient être désagréables au Divin. Il faut dire que
longtemps on n'a pas pris ces plaisanteries au sérieux parce que les
ennemis de l'Arétin ont inventé contre lui trop de calomnies pour que
ce qu'ils ont avancé soit admis sans discussion s'il n'est étayé par
des documents irréfutables. Mais, ne se manifestant pas avec
beaucoup de vivacité à l'égard de son père, l'amour filial de notre
Pierre se reporta tout entier sur sa mère, une très belle fille du
peuple nommée Tita. L'Arétin l'aima tendrement. On en a conclu
qu'elle était mariée. Et rien n'est moins certain. Messer Luca pouvait
bien vivre en concubinage avec Monna Tita. Elle a passé pendant
quelques siècles pour une mérétrice de bas étage et certains
arétiniens voudraient maintenant en faire une sainte! Il n'y a pas
apparence de cela. L'Arétin pouvait bien aimer sa mère de tout son
cœur, au cas même où elle eût été une prostituée. Au demeurant,
on n'est pas au courant de la vie que mena la Tite, mais on est
certain avant tout de sa beauté, dont furent touchés de nombreux
artistes qui voulurent la rendre immortelle.
En somme, l'origine de l'Arétin est obscure, mais nullement
monstrueuse. On est loin du sacrilège qui, lui donnant pour parent
un tertiaire et une béguine, faisait de lui l'Antéchrist même; selon la
légende encore accréditée qui veut que le père de cette incarnation
du mal, encore à venir, soit un religieux et sa mère une religieuse.
On sait aussi que l'Arétin eut deux sœurs qui se marièrent.
L'enfance de notre Pierre fut assez négligée. Il était précoce, lisait
tout ce qui lui tombait sous la main, dévorant avant tout avec
passion les romans chevaleresques, les divers épisodes épiques dont
sont formés i Reali di Francia, ces royaux de France dont plus tard il
devait combattre l'influence très considérable à cette époque en
Italie où ils ne sont pas encore oubliés aujourd'hui. Il alla bientôt à
Pérouse où, faisant déjà des vers, il étudia la peinture. Un livre
découvert à la Marciana par M. d'Ancona, en prouvant la précocité
poétique du jeune Arétin, démontre aussi qu'il se destinait pour les
arts: Opera nova del fecundissimo giovene Pietro Aretino zoé
strambotti, sonetti, capitoli, epistole, barzellette e una desperata; et
à la fin: Impresso in Venetia per Nicolo Zopino net MCCCCCXI a di
XXII di Zenaro. L'Arétin avait alors 19 ans. Les sonnets sont précédés
de cet avertissement: quelques choses d'un adolescent Arétin Pierre
étudiant en cette faculté et en peinture. Un sonnet dans lequel il est
question d'un Pérugin indique assez que l'Arétin était alors à
Pérouse. Un capitolo trouvé plus tard sur une colonne, à Rialto, en
novembre 1532, fait aussi allusion à ces tentatives artistiques:
O combien cela t'aurait rapporté plus de fruit et de louange
Si tu n'avais pas laissé ton pinceau,
S'il est vrai que tu aies été peintre un temps, comme je l'ai entendu
dire,
Plutôt que de vouloir devenir, ô petit misérable,
De Maître, poète.

En 1517, l'Arétin alla à Rome. Il y fut vite connu et craint à cause de


ses satires. Il entra au service du pape Léon X et du cardinal Jules
de Médicis. Après avoir fait une violente opposition à l'élection
d'Adrien VI, le détesté pape flamand, en prenant pour interprètes
Marforio et Pasquin, l'Arétin quitta Rome avec le cardinal et ne revint
que lorsque celui-ci fut élu pape sous le nom de Clément VII, le 19
nov. 1523. L'Arétin avait alors 31 ans. Il jouissait à la cour de
Clément de beaucoup de considération et pouvait beaucoup sur
l'esprit du pontife[9].
En 1524 éclate le scandale des figures de Jules Romain, gravées par
Marc Antoine. En 1525, l'Arétin écrit les 16 sonnets. Il est en guerre
avec le Dataire Giberti, qui tente de le faire assassiner par le
Bolonais Achille de la Volta. A peine remis de ses blessures, Messer
Pietro quitte Rome pour aller retrouver Jean des Bandes-Noires qui
l'accueille à bras ouverts. Le fameux capitaine meurt en 1526.
L'Arétin, revenu à Rome, assiste au sac de la ville. Clément VII
meurt, et l'Arétin, ne se sentant plus en sûreté, se réfugie à Venise,
où il arrive le 25 mars 1527, et s'y établit, disant aux cours un adieu
définitif. C'est alors qu'homme libre par la grâce de Dieu, il s'intitule:
le Fléau des Princes, le Véridique et le Divin. «Pourquoi, s'est
demandé Jacobus Gaddius[10], s'arrogea-t-il la divinité avec le
consentement de ses contemporains? Je ne sais. A moins que peut-
être il ne voulût signifier qu'il exerçait les fonctions de Dieu, en
foudroyant, au semblant de très hautes montagnes, les têtes les plus
élevées.»
A Venise, l'Arétin trouve le moyen de s'enrichir en écrivant des
lettres. Passant, tour à tour, du parti de François Ier dans celui de
Charles-Quint, respecté par le Roi et par l'Empereur, honoré par les
papes, l'Arétin, comblé d'honneurs, dispose de la plus haute
puissance de son temps. On le craint, on le flatte, il a de nombreux
ennemis dont il est à l'abri, et ses amis sont plus nombreux encore.
Ils font partie de toutes les classes de la société. Son nom est
fameux jusqu'en Perse. Il habite, sur le Canale Grande, un palais
somptueux détruit aujourd'hui. Au lieu d'intendant et de majordome,
ce sont six belles filles qui dirigent sa maison; on les appelle les
Arétines. Il choisit ses maîtresses comme ses commensaux, dans la
noblesse aussi bien que parmi le peuple. Sa maison est ouverte à
tous comme un port de mer. C'est une hôtellerie pour les pèlerins
affligés, pour les lettrés affamés et pour toute sorte de chevaliers
errants. Généreux à l'excès, il donne ce qu'il possède, ne parvenant
pas cependant à s'appauvrir. Chaque jour, de sa petite écriture nette
et nerveuse, il écrit des lettres destinées, par menaces ou par
flatteries, à provoquer des dons, à entretenir l'admiration et une
sainte terreur de sa plume étincelante. Il écrit vite, improvisant, en
quelque sorte, des comédies où l'a pu voir en lui un précurseur de
Molière, des écrits satiriques et libres selon la mode du temps, des
paraphrases religieuses pour lesquelles il doit ambitionner en vain le
chapeau de cardinal. Il compose des poèmes chevaleresques qui
n'en finissent plus et qu'il détruit lui-même, mais pour se consoler en
écrit des parodies. L'influence de ces faciles écrits se fait sentir non
seulement en Italie, mais en France, en Espagne, en Allemagne. Il
règle le goût, s'intéresse aux artistes et entasse chez lui les œuvres
d'art.
A peine à Venise, il rencontre le Titien, qui devient son compère, et
commence immédiatement son premier portrait qui, trois mois
après, fut envoyé au marquis de Mantoue. L'amitié du peintre et du
Divin ne devait plus cesser. Parmi ses amis on peut citer encore le
Sansovino, Sébastien del Piombo, le Sodoma, Jules Romain, Giovanni
da Udine et même Michel-Ange qui, s'il semble n'avoir jamais voulu
donner de ses œuvres à l'Arétin, qui sollicitait ce don, n'en tenait pas
moins le Fléau des Princes en haute estime, écrivant: «Le Roi et
l'Empereur avaient en très grande grâce que la plume de l'Arétin les
nommât.»
Dans le palais qu'il habitait se pressait chaque jour la foule des
artistes, des disciples, des patriciens, des aventuriers, des
ecclésiastiques, des mérétrices, des ganymèdes et des étrangers.
L'Arétin plaisante et rit souvent à gorge déployée. Il est l'homme le
plus libre du monde, il ne craint personne. Il reçoit des présents de
tous les souverains. François Ier et Charles-Quint lui ont donné des
chaînes d'or mais ne l'ont point enchaîné. Il se croit le droit de
changer de parti. Il a conscience de sa puissance. Et, seul parmi les
gens de lettres de son temps, il n'est pas parasite. On a dit que
c'était un maître-chanteur, mais on a exagéré. Il a des talents et
peut rendre des services. Il n'est que trop juste qu'on les lui paye. Il
ne ménage rien et dit hardiment sa pensée. Il a reproché au roi de
France d'avoir, à cause de son alliance avec les Turcs, plongé dans le
cœur de la chrétienté le couteau ottoman. Fléau des Princes, il les
flagelle par droit divin. L'opinion publique lui était, après tout, très
favorable, et les prédicateurs ne se gênaient pas pour déclarer que,
poursuivant le dessein de réformer la nation humaine, la nature et
Dieu ne pourraient pas trouver de meilleur moyen que de produire
beaucoup de Pierre Arétin.
Le Divin ayant quitté les cours en a maintenant une dans laquelle il
se promène en despote bon enfant, incapable de maîtriser ses
colères sans durée, et bon de cette bonté qui faisait dire à Jean des
Bandes-Noires qu'elle était la source de la plupart des désagréments
éprouvés par Messer Pietro. Et, de fait, il veut que tout le monde soit
heureux autour de lui. Pour cela il est très humain avec les femmes
de sa maison, jovial, hospitalier et généreux, tenant table ouverte,
libéral au point de donner cela même à quoi il tient le plus. Le
regard du Divino va de la vue merveilleuse qu'on découvre de son
palais au groupe des joueurs, aux artistes qui disputent sur l'idéal, il
s'arrête avec complaisance sur les belles courtisanes, sur les
honnêtes dames et sur les ganymèdes aux formes lascives. Car s'il
aime beaucoup les femmes et si deux fois au moins il a connu le
véritable amour qui est passionné, respectueux et même sans espoir,
il ne méprise pas des plaisirs qui, comme aujourd'hui même,
choquant l'autorité, ne passaient pour honteux qu'aux yeux d'un très
petit nombre de particuliers. Il ne faut pas oublier que
Giovannantonio Bazzi n'a pas peur d'être appelé le Sodoma, que le
Berni, le Tasse, Michel-Ange et bien d'autres eussent mérité le même
surnom. Mais pour l'amant de Laura la cuisinière, de la comtesse
Matrina, de la vertueuse Angelo Serena, de la malheureuse et frivole
Perinia Riccia, le caprice socratique n'a que l'importance passagère
d'un divertissement. Il a des filles et s'occupe de leur établissement.
Le Divino, que l'Arioste a célébré, que François Ier, charmé par son
esprit, avait voulu attirer à sa cour, que Charles-Quint fit chevaucher
à son côté, que le pape Jules III baisa au front et auquel il conféra
l'ordre de Saint-Pierre, eut une vieillesse magnifique, et l'Ammirato
dit qu'on aurait difficilement vu un vieillard plus beau et plus
pompeusement vêtu. Les fables les plus grossières ont couru sur les
circonstances qui entourèrent la mort du Flagello dei principi. On a
retrouvé un témoignage authentique et précis de son décès. C'est un
certificat notarié et revêtu du firman ducal fait à la requête d'un
certain Domenico Nardi da Reggio, probablement pour couper court
aux bruits calomnieux qui commençaient à courir sur la mort de
l'Arétin. Il contient les déclarations de Pietro Paolo Demetrio, curé de
San Luca, paroisse du Divin, à Venise. Ce curé atteste, en 1581,
c'est-à-dire 25 ans après la mort de Pierre, avoir enseveli
chrétiennement l'Arétin et dit qu'il mourut de mort subite, tombant
d'une chaise caquetoire, et que le jeudi saint avant de finir ses
ultimes jours il se confessa et communia, pleurant extrêmement, et
le bon prêtre affirme que cela s'est bien passé ainsi comme il l'a vu
lui-même.
C'est que l'Arétin n'était pas un mécréant. Il avait un confesseur, le
père Angelo Testa, et suivait les offices. S'il se moque des moines, il
respecte infiniment la religion. Jules III n'a pas voulu en faire un
cardinal. Et ce refus me paraît avoir eu des raisons plus politiques
que morales. L'Arétin était, autant que bien d'autres, digne de la
pourpre cardinalice et n'aurait peut-être pas fait si mauvaise figure
sur le trône pontifical!
L'Arétin a laissé une œuvre importante[11]; outre ses lettres
laudatives, ses pamphlets et ses poésies de circonstance, il a donné
une tragédie en vers, Orazia, et cinq comédies en prose: Le
Maréchal, la Courtisane, l'Hypocrite, la Talanta, le Philosophe, où l'on
découvre des mérites du premier ordre. On a bien avancé que
l'Hypocrite aurait été le prototype du Tartufe, Molière ayant connu
cette pièce à Grenoble, grâce à Chorier[12]. Les ouvrages religieux
du Divin eurent une vogue considérable. Il paraphrase les psaumes
pénitentiels, parle de l'Humanité du Christ, de la vie de la Vierge
Marie, de la Passion de Jésus-Christ, de la vie de Sainte Catherine. Il
a composé une œuvre chevaleresque dont les strophes se
comptaient par dizaines de mille, mais il la détruisit lui-même, ne
nous laissant que des poèmes inachevés comme le Lagrime
d'Angelica ou la Marfisa et des parodies également inachevées,
comme l'Orlandino qui eut un très grand succès, et l'Astolfeide dont
on ne connaît qu'un exemplaire conservé à la Bibliothèque Nationale
et sur lequel on trouve cette note manuscrite: Non ce ne sono che
Tre Canti. Molte Coglionerie, e pochissime cose.
On a dit de l'Arétin qu'il était un grand prosateur, mais un poète
médiocre. Je suis d'avis que cette opinion est en partie très injuste,
car le Divin a été pour le moins un poète satirique du premier ordre.
Certaines de ses pasquinades[13] ne sont pas inférieures à quelques
beaux morceaux de Victor Hugo, dans les Châtiments.
Pour ma part, je suis d'avis que l'on devrait restituer à l'Arétin la
paternité de quelques ouvrages comme la Puttana errante[14], la
Zaffetta, la Tariffa delle Puttane que l'on attribue à Lorenzio Veniero.
Ce Lorenzo Veniero, qui devait plus tard siéger au Sénat et remplir
de hautes fonctions dans le gouvernement de la République
Vénitienne, avait vingt ans lorsque Francesco Zeno l'amena à l'Arétin
pour que celui-ci le formât. Et ma conviction est faite: la Puttana
errante, la Zaffetta et son Trentuno ont trop de points de
ressemblance avec les Ragionamenti pour qu'il soit possible de les
attribuer à un autre qu'à l'Arétin lui-même. Je pense que le Divin ne
se souciait pas de s'attirer des désagréments en se moquant
ouvertement des mérétrices. Il avait sans doute à se venger de cette
Elena Ballerina, qui est la putain errante, et de la Zaffetta. Il a plu à
l'Arétin de mettre ses sarcasmes sur le compte du jeune Veniero, qui
ne demandait pas mieux et qui, sans doute, était très fier de se faire
passer pour l'auteur d'ouvrages d'une audace aussi brillante. Et,
cependant, l'Arétin a beau dire que la Puttana est l'œuvre du
Venerio, son creato, il a beau, au début de la Zaffetta, parlant au
nom du Veniero, se gausser de ceux qui disent que la Puttana
errante est un ouvrage arétinesque; il ne faut pas se laisser prendre
à ces supercheries et à ces coquetteries d'auteur. Au fond, l'Arétin
regrette d'avoir dépensé tant d'esprit dont bénéficie son disciple, il
reprend les traits les mieux venus de ses poèmes et s'en ressert
dans les Ragionamenti, y mentionnant La Putain errante en se
gardant bien de parler du Venerio. Le Tarif des putains de Venise
ressemble trop à la Putain errante et à la Zaffetta pour ne point
provenir de la même imagination. Cette composition, dont le titre
italien est La Tarifa delle Puttane di Venegia, a été écrite sans doute
entre la première et la deuxième partie des Ragionamenti. L'Arétin la
mentionne dans la première journée de cette deuxième partie. Il la
fit probablement paraître plus tard, y ayant mis des allusions à lui-
même et au Veniero pour qu'on ne découvrît pas quel en était
l'auteur.
Bref, si l'Arétin n'a pas écrit les trois ouvrages dont il a été question,
il leur a beaucoup emprunté, et cela n'est pas dans ses habitudes. Il
tire, en général, de son propre fonds tout ce qu'il écrit. Il travaille si
vite que plagier ne pourrait que le retarder inutilement. D'ailleurs,
n'a-t-il pas dit dans une phrase qu'on pourrait rapprocher d'un vers
de Musset: «Il vaut mieux boire dans son hanap de bois que dans la
coupe d'or d'autrui.»
Je ne veux nullement avancer, au demeurant, que l'Arétin, qui était
presque un autodidacte, n'ait pas subi l'influence d'auteurs qui l'ont
précédé ou même contemporains. Sans parler de Boccace et des
autres Italiens dont la lecture a formé son esprit en lui donnant une
direction, il serait injuste de ne pas citer l'Espagnol Francisco
Delicado qui paraît avoir eu une influence immédiate sur le talent du
Divin. Ce Francisque ou François Délicat, dont la vie, le rôle et les
œuvres sont encore mal connus, vivait en Italie. Il était à Rome en
même temps que l'Arétin et alla à Venise la même année que lui. Il
publia, en 1528, avant que l'Arétin ne composât ses Journées
putanesques, une nouvelle dramatique intitulée La Lozana Andaluza,
qui pourrait bien être le prototype des Ragionamenti, ayant elle-
même pour mobile la fameuse Célestine. L'Arétin entendait
l'espagnol, comme il apparaît à la lecture de ses dialogues. Il a dû
connaître La Lozana Andaluza et sans doute son auteur, qui était un
lettré et un savant. Quoi qu'il en soit, il ne le mentionne nulle part.
La Lozana Andaluza fut composée à Rome pendant le séjour qu'y fit
Délicat, de 1523 à 1527. Il la retoucha à Venise avant de l'y publier.
J'attribuerais volontiers à ce Francisque Délicat un ouvrage qui a été
longtemps donné comme étant de l'Arétin et qui a comme titre le
nom d'un fameux éditeur vénitien. Je veux parler du Zoppino, dans
lequel on reconnaîtra volontiers bien des traces du goût espagnol.
En tout cas, le Zoppino n'est pas de l'Arétin, tout le monde est
d'accord à ce sujet. D'autre part, au Mamotreto ou cahier XXXIX de
la Lozana Andaluza, Délicat mentionne le Zoppino qui ne devait
paraître à Venise qu'en 1539, après les Six Journées ou Caprices de
l'Arétin. Et l'on trouverait bien des ressemblances entre la Lozana
Andaluza et le Zoppino qui tous deux, sans doute, furent composés à
Rome et retouchés à Venise. Délicat devait écrire l'italien, et dans
son séjour à Venise il se mit au courant du dialecte vénitien auquel il
a emprunté un certain nombre de locutions qui paraissent dans le
Zoppino. Il ne cite pas une fois l'Arétin, sans doute parce que celui-ci
ne l'avait pas cité non plus. Il intitule son dialogue: Ragionamento
del Zoppino, etc., imitant en cela l'Arétin, à moins que celui-ci
n'ayant connu le Zoppino à Rome n'en ait imité le titre avant qu'il ne
fût imprimé.
Néanmoins, l'Arétin échappe, quant à son ouvrage même des
Caprices, à tout reproche d'imitation et de plagiat, de même que
Francisque Délicat ne peut être appelé un imitateur de la Célestine,
bien qu'elle ait été le modèle de la Lozana Andaluza dont elle diffère
de toutes les façons. Mes hypothèses sur l'influence et les ouvrages
de Francisque Délicat n'infirment point, du reste, mes opinions
touchant la Putain, la Zafetta et le Tarif qui me semblent devoir être
remis au compte de l'imagination féconde du Divin. Il ne s'est caché
de les avoir écrits que parce qu'à Venise, attaquer nommément la
renommée des mérétrices de la République et même des courtisanes
romaines, cela pouvait être infiniment plus dangereux que de se
moquer du roi de France, et surtout cela ne devait rien rapporter.

On a pensé que le Divin, dont le nom est populaire en France, y était


trop mal connu, et l'on a choisi pour le faire connaître les ouvrages
dans lesquels sa personnalité s'est affirmée le plus et qui lui font une
place à part parmi les écrivains de tous les temps. On n'a donné ici
que les seize Sonnets luxurieux qui paraissent être de l'Arétin. On
sait que ces sonnets ont été portés jusqu'à 26, nombre qui ne
répond pas à celui des figures de Jules Romain.
Il n'existe pas encore de travail définitif touchant l'histoire de ces
sonnets; néanmoins celui[15] du savant Alcide Bonneau, à l'érudition
élégante et inépuisable duquel on doit la plupart des travaux publiés
par Liseux, fait autorité. Pour ce qui a trait aux fameux dessins de
Jules Romain, gravés par Marc-Antoine Raimondi, ils ont
complètement disparu. On a donné récemment une réimpression
des sonnets, copiée sur l'édition de Liseux. On y a ajouté les fac-
similés d'une série de calques datant du xviiie siècle et qui auraient
été faits sur les gravures de Marc-Antoine[16]. Mais n'y a-t-il pas là-
dessous quelque supercherie? Ces images coïncident presque
entièrement avec la description qu'avait donnée Bonneau de
l'apparence que devaient avoir les gravures disparues. Mais sont-ce
bien là des calques datant du xviiie siècle ou bien ne s'agirait-il pas
plutôt d'une habile reconstitution faite d'après la description de
Bonneau et où l'on a mis quelques différences pour que
l'authenticité des calques parût moins discutable? Je ne sais.
Toujours est-il que cette publication a été saisie après son apparition
et son éditeur poursuivi.
On ne comprend pas bien dans ces conditions pourquoi la
Bibliothèque nationale n'en possède pas un exemplaire. Sans doute,
l'institution du Dépôt légal ne fonctionne pas avec toute la régularité
désirable; mais un ouvrage ayant été saisi, le premier geste de
l'autorité devrait être d'en pourvoir la Bibliothèque, dont on se
désintéresse trop. On dit que les magistrats, en cas de saisie comme
celle dont il est question ici, s'empressent de compléter leurs
collections. Et sans doute il y a trop de collectionneurs dans la
magistrature pour que d'un ouvrage saisi il ne reste un seul
exemplaire destiné à la Nationale.
On a dit que l'éditeur était parvenu à se faire rendre son édition.
Cependant, je crois qu'elle ne lui a pas été rendue, mais qu'il en a
tiré une nouvelle, les exemplaires que l'on vend maintenant me
paraissant plus petits et moins beaux que ceux que j'ai vus en 1904.
Néanmoins, je ne pourrais pas affirmer le fait, parce qu'en 1904, ne
m'occupant pas encore de l'Arétin, je n'ai pas regardé avec
beaucoup d'attention la publication qui venait de paraître.
En se servant du recueil du Cosmopolite[17], Alcide Bonneau a pu
reconstituer avec beaucoup de vraisemblance l'ouvrage fescennin du
Divin. Ce n'est pas que parmi les autres sonnets il n'y en ait pas qui
puissent être aussi attribués à l'Arétin. Ainsi le sonnet qui sert de
préambule à la Corona de Cazzi, comme on a appelé
postérieurement les Sonnets luxurieux, peut fort bien être également
de l'Arétin. Le premier quatrain est aussi le premier du sonnet qui
sert de poème à la Tariffa delle Puttane di Venegia, que, pour ma
part, j'attribue à l'Arétin.
Pour ce qui regarde les Ragionamenti, on a traduit ici la première
partie qui se compose de trois Journées. Il y manque l'Avertissement
dans lequel l'Arétin dédie son ouvrage à sa guenon en jouant sans
doute sur le mot mona[18] qui avait à Venise un autre sens que l'on
entend assez si l'on a parcouru les priapées que le Vénitien Baffo
composa au xviiie siècle. La troisième Journée est la plus célèbre.
Dès le xvie siècle, elle était imitée plutôt que traduite en français, et
aussi en espagnol (1549). C'est d'après cette paraphrase intitulée
Colloquio de las Damas et due à Fernand Xuarès que Gaspard Barth
composa sa fameuse traduction latine intitulée Pornodidascalus.
La seconde partie est également formée de trois Journées qu'Alcide
Bonneau a respectivement intitulées: l'Éducation de la Pippa, les
Roueries des Hommes, la Ruffianerie. Dans la première de ces
Journées, la Nanna enseigne à sa fille, la Pippa, l'art d'être
mérétrice. Le second jour, il s'agit des bons tours que les hommes
s'ingénient à jouer aux courtisanes trop confiantes. Et le troisième
jour, la Nanna et la Pippa, assises dans leur jardin, écoutent la
Commère et la Nourrice parler de la Ruffianerie, c'est-à-dire des
rapports entre les putains et les maquerelles. On a souvent donné le
Zoppino, le Ragionamento des Cours et le Dialogue du Jeu comme
étant la troisième partie des Ragionamenti. C'est là une erreur. Le
Zoppino n'est pas de l'Arétin et les Six journées forment une œuvre
distincte et complète. Le Ragionamento des Cours n'a pas encore
été traduit; il mérite cependant de l'être. Quant au Dialogue du Jeu,
on en a traduit des fragments, et il n'est pas indigne non plus qu'on
en publie une version complète.
Les traductions que l'on donne ici paraîtront souvent plus exactes
que celles qui les ont précédées. Le traducteur de l'édition de Liseux,
malgré tous ses mérites, n'a pas évité quelques contresens
regrettables comme celui-ci au deuxième dialogue où il traduit
spazzare ogni gran camino par «balayer la poussière des plus larges
chemins». Ce qui n'était évidemment pas ce que voulait dire le Divin,
les ramoneurs étant de son temps plus communs que les
cantonniers. On a aussi serré le texte italien de plus près. C'est ainsi
qu'on a rendu schiavina, non pas seulement par «manteau», mais
par «esclavine», et que traduire le fu renduto da me migliaccio per
torta par «je lui rendis mille pour un» a paru une étrange façon de
faire passer dans l'officine de l'usurier une locution populaire qui
sortait sans doute du fournil du boulanger. On n'a pas reculé non
plus devant les répétitions que n'avait pas évitées l'Arétin qui écrivit
ses Ragionamenti en 48 jours. Il a paru que l'office du traducteur ne
doit pas être d'améliorer le style de son auteur, et l'on n'est pas
éloigné de croire, au demeurant, que les répétitions ne sont
nullement un indice de mauvais style comme on pense
communément aujourd'hui, où l'on alourdit et embarrasse souvent la
phrase en voulant se servir de mots différents là où la répétition d'un
mot serait aussi bien raisonnable.
Enfin, on a mis des notes partout où cela a été possible. On souhaite
qu'elles éclaircissent un texte très agréable à la vérité, mais rempli
d'allusions à des événements, à des coutumes, à des personnages
dont le public n'a pas idée aujourd'hui.
En ce qui concerne les sonnets, on en a parfois adouci les termes, et
malgré cela on est persuadé que ces poèmes n'ont pour ainsi dire
rien perdu de leur vivacité gaillarde. D'ailleurs, le lecteur est libre de
remplacer les mots qui lui paraissent faibles par les plus forts qu'il
connaisse, et suppléant ainsi par la perspicacité de son entendement
à ce que le traducteur a dû gazer, par pudeur, il formera avec
certitude son opinion sur l'œuvre du Divin Pierre Arétin dont on a
écrit en son temps qu'il était la règle de tous et la balance du style.
GA
[1] Si l'on a pu citer Rabelais et Molière comme des auteurs sur lesquels le Divin a
exercé son influence, il serait injuste de ne pas ajouter que, de notre temps,
Hugues Rebell, qui était un grand lecteur des publications de Liseux, a dû à l'Arétin
une très grande partie de ses mérites d'écrivain.
[2] Toutes les nuances des attitudes galantes ont été traitées avec «tant d'énergie
par le célèbre Pierre Arétin, qui vivait dans le quinzième siècle (sic), qu'il n'en reste
rien à dire aujourd'hui». Thérèse philosophe, 2e partie. Cette opinion, exprimée
dans un des ouvrages les plus licencieux du xviiie siècle, représente bien l'idée que
l'on se fait encore en général du Divin.
[3] Cette traduction fut d'abord publiée sur le texte italien en dix volumes (1879-
1880). Petite édition mixte franco-latine.
J'ai eu entre les mains une traduction très rare, mais peu intéressante. Il s'agit des
Dialogues de l'Arétin surnommé le fléau des Princes, le véridique, le divin. Paris,
1884, 4. vol. in-8º. Cet ouvrage a été imprimé sur la presse à bras par le
traducteur A. Ribeaucourt et tiré à 15 exemplaires seulement.
[4] L'ouvrage suivant a fait longtemps autorité: Vita di P. Aretino; par le comte G.-
M. Mazuchelli (Padoue, 1741, 1749). Il y en a un abrégé en français, par Dujardin,
sous le pseudonyme de Boispréaux (La Haye, 1750). On trouve bien quelques
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and


personal growth!

textbookfull.com

You might also like