Reading Group

The mailman list for the reading group is “ml-security”. To subscribe, visit:

For Fall 2021, we will be meeting on Thursdays 5:00-6:30. We will start off virtually and then hopefully, with the case numbers coming down, we will move to in-person.

DatesTopicSuggested papersPresentersRecording
Sep 6Introductions, Fun quiz, Paper writing tips

Sep 16Own work: Ensemble security for MLRuqi
Sep 23, 30Reliability of zero/one shot learningA Generative Adversarial Approach for Zero-Shot
Learning from Noisy Texts (CVPR, 2018)
AshrafRecording (protected)
Slide deck
Oct 7Model extraction attacks and defensesTramèr, Florian, Fan Zhang, Ari Juels, Michael K. Reiter, and Thomas Ristenpart. "Stealing machine learning models via prediction apis." In 25th {USENIX} Security Symposium ({USENIX} Security 16), pp. 601-618. 2016.Ahaan, AtulRecording (protected)
Slide deck
Oct 14October break
Oct 21, 28Security of distributed learningLiu, Lumin, Jun Zhang, S. H. Song, and Khaled B. Letaief. "Client-edge-cloud hierarchical federated learning." In ICC 2020-2020 IEEE International Conference on Communications (ICC), pp. 1-6. IEEE, 2020.
Wang, Xiaoding, Sahil Garg, Hui Lin, Jia Hu, Georges Kaddoum, Md Jalil Piran, and M. Shamim Hossain. "Towards accurate anomaly detection in industrial internet-of-things using hierarchical federated learning." IEEE Internet of Things Journal (2021).
Edgardo, Josh ZSlide deck
Nov 4Own workAkhilVideo
(Password protected)
Nov 11Large streaming models on tiny devices"Flexible High-resolution Object Detection on Edge Devices with Tunable Latency" MSRA, USTC, Tsinghua, Mobicom 2021.
"MoViNets: Mobile Video Networks for Efficient Video Recognition" Google Research, CVPR 2021.
Pengcheng, PreetiPengcheng: Mobicom 2021
Preeti: CVPR 2021
Nov 18Own work: Approximate streaming models on embedded devicesAkash, Sarthak
Nov 25Thanksgiving break
Dec 2Own workMustafa
Dec 9Adversarial reinforcement learningPinto, Lerrel, James Davidson, Rahul Sukthankar, and Abhinav Gupta. "Robust adversarial reinforcement learning." In International Conference on Machine Learning, pp. 2817-2826. PMLR, 2017.
Gleave, Adam, Michael Dennis, Cody Wild, Neel Kant, Sergey Levine, and Stuart Russell. "Adversarial policies: Attacking deep reinforcement learning." ICLR 2020.
Video (protected)

For Spring 2021, we will be meeting on Tuesdays 11.30-1.00.

The papers are samples and the discussion will go beyond them or may use other papers altogether.

Jan 26, Feb 2How to write a strong evaluation sectionSaurabh
Feb 9Lightning talks from all group membersAll
Feb 16Own work: Security of Android devicesJosh
Feb 23, Mar 2Reliability and security of serverless computing
Sreekanti et al. "A Fault-Tolerance Shim for Serverless Computing" Eurosys 2020.
Datta et al. "Valve: Securing Function Workflows on Serverless Computing Platforms" WWW 2020.
Karthick, Ashraf
Mar 9Own work: IoT deploymentsPengcheng, Edgardo
Mar 16, 23Reliable computing with heterogeneous embedded devices and dronesRan, Jay
Mar 30Own work: Security in distributed MLAtul
Apr 6, 20 (Apr 13 is reading day and off for students)Distributed ML on lightweight devicesPranjal, Tanushree
Apr 27Own work: Security in MLRuqi
May 4Lessons learnedSaurabh

History of DCSL Reading group

2016-20, 2015, 2014, 2013, 2012, 2010, 2009, 2008, 2007, 2006, 2005, 2004, 2003, 2002

Last modified: September 6, 2021