SPSL Call for Papers
https://sites.google.com/usc.edu/spsl/home
The first workshop on Secure and Private Systems for machine Learning (SPSL) will be co-located with ISCA. This workshop is meant to serve as a platform to promote discussions on system-level approaches for secure and privacy-preserving machine learning by bringing together researchers from computer systems, security, and machine learning. This one day virtual workshop will be co-located with ISCA and is scheduled for Jun 12th 2021.
Topics of submitted papers include, but not limited to, the following areas, with particular emphasis on system design and implementation aspects:
-
Systems for privacy-preserving training and/or inference
-
Systems for robustness of ML computations
-
Secure hardware (such as trusted execution environment) for secure and private ML
-
System support and acceleration for algorithmic solutions such as secure multi-party computation (MPC), homomorphic encryption (HE), federated learning, and differential privacy
-
Hardware-algorithm co-design for ML security and/or privacy
-
System-level challenges in applied cryptography for ML
-
New attacks and defenses on ML systems, including side-channel attacks.
-
Study and evaluation of practical ML system security and privacy
MPORTANT DATES
April 23rd, 2021: Abstract submission deadline
April 30th, 2021: 4-page Paper submission deadline
May 3rd, 2021: Paper assignments to PC
May 26th, 2021: Paper reviews due
May. 26th – June 1st, 2021: Online discussion for paper selection
June 1st 2021: Paper notifications to authors
Organizing Committee
-
Murali Annavaram (University of Southern California)
-
Edward Suh (Cornell University, Facebook AI Research)
-
Wenjie Xiong (Facebook AI Research)
-
Hanieh Hashemi (University of Southern California)