CS492: Machine Learning Application Trends in Information Security

The course is a paper reading class. The class will cover a list of papers published at the prestigious security conferences. Each student will present assigned papers and lead discussions. The goal of the course is to understand the trends in applying machine learning algorithms on computer security problems as well as to seek an in-depth understanding of the covered research papers.

Basic Information

• Lecture: Tuesday/Thursday 9:00 AM - 10:30 AM
• Instructor: Sooel Son
• Email: sl.son (at) kaist.ac.kr
• Homepage: https://sites.google.com/site/ssonkaist/
• Office: Room 2312 N5
• Office hours: Every Tuesday 10:30 AM to 11:50 AM
• T.A.: Suyoung Lee, suyoung.lee (at) kaist.ac.kr

Evaluation

• Attendance & Class participation: 15%
• Paper critiques: 15%
• Paper Presentation #1: 10%
• Paper Presentation #2: 10%
• Paper Presentation #3: 10%
• Project proposal & Midpoint evaluation : 10%
• Final project: 30%

Schedule

• 9/3 Course Introduction
  • Presenter: Sooel Son
  • Slide
• 9/5
  • [1] Zhou et al. Hardware Performance Counters Can Detect Malware: Myth or Fact? (ASIACCS 2018)
  • Presenter: Sooel Son
• 9/10
  • [2] Bilge et al. RiskTeller: Predicting the Risk of Cyber Incidents. (CCS 2017)
  • Presenter: Sooel Son
  • Slide
• 9/12
  • No class (Holiday)
• 9/17
  • [3] Xu et al. Neural Network-based Graph Embedding for Cross-Platform Binary Code Similarity Detection. (CCS 2017)
  • Presenter: Suyoung Lee
  • Slide
• 9/19
  • [4] When Coding Style Survives Compilation: De-anonymizing Programmers from Executable Binaries. (NDSS 2018)
  • Presenter: Sunnyeo Park
  • Slide
  • Previous Slide
• 9/24
  • [5] Song et al. Machine Learning Models That Remember Too Much. (CCS 2017)
  • Presenter: Corentin Soulet
  • Slide
  • Previous Slide
• 9/26
  • [6] Schuster et al. Beauty and the Burst: Remote Identification of Encrypted Video Streams (Usenix 2017)
  • Presenter: Ayoub Mellah
  • Project Proposal Deadline
  • Slide
  • Previous Slide
• 10/1
  • [7] Barradas et al. Effective Detection of Multimedia Protocol Tunneling using Machine Learning. (Usenix 2018)
  • Presenter: Sooel Son
  • Slide
  • Previous Slide

• 10/3 No class (Holiday)

• 10/8
  • [8] Shokri et al. Membership Inference Attacks Against Machine Learning Models. (S&P 2017)
  • Presenter: Joongyum Kim
  • Slide
  • Previous Slide
• 10/10
  • [9] Pygelis et al. Knock Knock, Who’s There? Membership Inference on Aggregate Location Data. (NDSS 2018)
  • Presenter: Farid Talibi
  • Slide
  • Previous Slide
• 10/15
  • [10] Pei et al. DeepXplore: Automated Whitebox Testing of Deep Learning Systems. (SOSP 2017)
  • Presenter: Sunnyeo Park
  • Slide
  • Previous Slide
• 10/17
  • [11] Tial et al. DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Cars. (ICSE 2018)
  • Presenter: Corentin Soulet
  • Slide
  • Previous Slide

• 10/22 Midterm season

• 10/24 Midterm season

• 10/29
  • [12] Calini et al. Towards Evaluating the Robustness of Neural Networks. (S&P 2017)
  • Presenter: Joongyum Kim
  • Slide
  • Previous Slide

• 10/31 Project Midterm Evaluation

• 11/5
  • [13] Meng et al. MagNet: a Two-Pronged Defense against Adversarial Examples. (CCS 2017)
  • Presenter: Ayoub Mellah
  • Slide
  • Previous Slide
• 11/7
  • [14] Yinzheng et al. Practical Attacks Against Graph-based Clustering. (CCS 2017)
  • Presenter: Farid Talibi
  • Slide
  • Previous Slide
• 11/12
  • [15] Ye et al. Yet Another Text Captcha Solver: A Generative Adversarial Network Based Approach. (CCS 2018)
  • Presenter: Suyoung Lee
  • Slide
  • Previous Slide
• 11/14
  • No class
• 11/19
  • [16] Wang et al. With Great Training Comes Great Vulnerability: Practical Attacks against Transfer Learning. (Usenix 2018)
  • Presenter: Corentin Soulet
  • Slide
  • Previous Slide
• 11/21
  • No class
• 11/26
  • [17] Suciu et al. When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attacks. (Usenix 2018)
  • Presenter: Joongyum Kim
  • Slide
  • Previous Slide
• 11/28
  • [19] DeepMem: Learning Graph Neural Network Models for Fast and Robust Memory Forensic Analysis. (CCS 2018)
  • Presenter: Sunnyeo Park
  • Slide
  • Previous Slide
  • [20] She et al. NEUZZ: Efficient Fuzzing with Neural Program Smoothing. (S&P 2019)
  • Presenter: Ayoub Mellah
  • Slide
• 12/3
  • [21] Melis et al. Exploiting Unintended Feature Leakage in Collaborative Learning. (S&P 2019)
  • Slide
  • Presenter: Farid Talibi
• 12/5
  • [22] Song et al. Auditing Data Provenance in Text-Generation Models. (KDD 2019)
  • Presenter: Sooel Son

• 12/10 Project presentation

• 12/12 Project presentation