Robustness and Security in ML Systems: Junfeng Yang

E6998 Robustness and Security in ML Systems

Spring 2019 -- Junfeng Yang

We put a tentative syllabus here to give you an idea what future may bring. This syllabus is subject to change as the course progresses.


# Day Date Topic Assignment Speakers
1 Mon Jan 28 Introduction Form reading group
2 Mon Feb 4 Deep learning Read Lecun-90c, AlexNet Robby, Chengzhi
3 Mon Feb 11 Adversarial ML (1) Read Intriguing properties of neural networks, Fast adversarial DL Shruti, Jonas
4 Mon Feb 18 Adversarial ML (2) Read Obfuscated gradients not useful, Blackbox adversarial DL Ziyuan, Chengzhi
5 Mon Feb 25 Interpreting DL (1) Read Extracting NN rules, NN visual representations Jake, Bryan
6 Mon Mar 4 Interpreting DL (2) Read Influence functions, NN interpretations fragile Hsiao-Yuan, Weiyu
7 Mon Mar 11 Verifying DL (1) Read Reluplex, DeepSafe Fan, Hongyi
8 Mon Mar 18 No class (Spring recess)
9 Mon Mar 25 Verifying DL (2) Read AI2, Abstract domain Ziyuan, Chengzhi
10 Mon Apr 1 Robustness training (1) Read Stability training, Robustness vs accuracy Robby, Yuchi
11 Mon Apr 8 Verifying DL (3) Read Reluval, Neurify Guest: Shiqi Wang
12 Mon Apr 15 Robustness training (2) Read Robust optimization, Robustness vs data Ziyuan, Chengzhi
13 Mon Apr 22 NN architecture Read Capsules, Gated graph NN Vinay, Abhishek
14 Mon Apr 29 Testing DL Read DeepXplore, VeriVis Guest: Kexin Pei
15 Mon May 6 Mini-research conference Present and demo your final project