Towards Evaluating the Security of Human Computable Passwords Using Neural Networks

Issei Murata, Pengju He, Yujie Gu, Kouichi Sakurai

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Passwords are playing a major role for authentication in our daily life. However contemporary passwords are typically either difficult to remember or vulnerable to various attacks. In 2017, Blocki, Blum, Datta and Vempala introduced the concept of human computable passwords as a promising authentication method. The fundamental concerns for designing human computable passwords are their usability and security. So far, the security evaluation on human computable passwords authentication schemes is mainly based on complexity-theoretic analysis. In this paper, we initially investigate the security of human computable passwords against neural network-based adversarial attacks. Specifically, we employ the typical multilayer perceptron (MLP) model to attempt to attack the human computable passwords authentication scheme proposed by Blocki-Blum-Datta-Vempala. We present implementation results and the corresponding analysis as well. Our results imply that it is possible for an MLP to learn a simple function, but is difficult for an MLP to learn piecewise functions well.

Original languageEnglish
Title of host publicationInformation Security Applications - 23rd International Conference, WISA 2022, Revised Selected Papers
EditorsIlsun You, Taek-Young Youn
PublisherSpringer Science and Business Media Deutschland GmbH
Pages303-312
Number of pages10
ISBN (Print)9783031256585
DOIs
Publication statusPublished - 2023
Event23rd International Conference on Information Security Applications, WISA 2022 - Jeju Island, Korea, Republic of
Duration: Aug 24 2022Aug 26 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13720 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference23rd International Conference on Information Security Applications, WISA 2022
Country/TerritoryKorea, Republic of
CityJeju Island
Period8/24/228/26/22

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Towards Evaluating the Security of Human Computable Passwords Using Neural Networks'. Together they form a unique fingerprint.

Cite this