Efficient RL-based Cache Vulnerability Exploration by Penalizing Useless Agent Actions

Kanato Nakanishi, Soramichi Akiyama

Presentation: In-Person

Keywords: Reinforcement Learning, Side-Channel Attack, Cache-Timing Attack

Presenter Full Name: Kanato Nakanishi

TL;DR: We propose a RL-based cache vulnerability exploration approach that reduces useless agent actions and reduces training time by up to 28%.

Presenter Email: [e-mail]

Abstract: Cache-timing attacks exploit microarchitectural characteristics to leak sensitive data, posing a severe threat to modern systems. Despite its severity, analyzing the vulnerability of a given cache structure against cache-timing attacks is challenging. To this end, a method based on Reinforcement Learning (RL) has been proposed to automatically explore vulnerabilities for a given cache structure. However, a naive RL-based approach suffers from inefficiencies due to the agent performing actions that do not contribute the exploration. In this paper, we propose a method to identify these useless actions during training and penalize them so that the agent avoid them and the exploration efficiency is improved. Experiments on 17 cache structures show that our training mechanism reduces the number of useless actions by up to 43.08% and the training time by up to 28% compared to a naive RL-based approach.

Presenter Bio: Kanato Nakanishi is a 2nd year master student at College of Information Science and Engineering, Ritsumeikan University, Japan. He is interested in Side-channel Attack, CPU Security, and Machine Learning.

Paper Checklist Guidelines: I certify that all co-authors have validated the presented results and conclusions, and have read and commit to adhering to the Paper Checklist Guidelines, Call for Papers and Publication Ethics.

YouTube Link: TBD

YouTube Link Poster: N/A

Google Slides: https://docs.google.com/presentation/d/17Q_muaP42wgQT2SHxNh1KRr0xTcCDm2Pu0pIaWAcN5k/edit?usp=sharing

Poster: Yes

Workshop Registration: Yes, the presenter has registered for the workshop.

YouTube Link Short: TBD

Submission Number: 8