02/02/2021

Differentially Private Stochastic Coordinate Descent

Georgios Damaskinos, Celestine Mendler-Dünner, Rachid Guerraoui, Nikolaos Papandreou, Thomas Parnell

Keywords:

Abstract Paper Similar Papers

Abstract: In this paper we tackle the challenge of making the stochastic coordinate descent algorithm differentially private. Compared to the classical gradient descent algorithm where updates operate on a single model vector and controlled noise addition to this vector suffices to hide critical information about individuals, stochastic coordinate descent crucially relies on keeping auxiliary information in memory during training. This auxiliary information provides an additional privacy leak and poses the major challenge addressed in this work. Driven by the insight that under independent noise addition, the consistency of the auxiliary information holds in expectation, we present DP-SCD, the first differentially private stochastic coordinate descent algorithm. We analyze our new method theoretically and argue that decoupling and parallelizing coordinate updates is essential for its utility. On the empirical side we demonstrate competitive performance against the popular stochastic gradient descent alternative (DP-SGD) while requiring significantly less tuning.

The video of this talk cannot be embedded. You can watch it here:
https://slideslive.com/38947885
(Link will open in new window)
 0
 0
 0
 0
  Share    
This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment
no comments yet
code of conduct: tbd Characters remaining: 140

Similar Papers

 0
 0
 0
 0
 5:09 
 0
 0
 0
 0
 13:08 
 0
 0
 0
 0
 4:56 
 0
 0
 0
 0
 3:09 
 0
 0
 0
 0
 3:13 
 0
 0
 0
 0
 19:06 
 0
 0
 0
 0
 3:02 
 0
 0
 0
 0
 3:23 
 0
 0
 0
 0
 15:09 
 0
 0
 0
 0
 15:38 
 0
 0
 0
 0
 5:24 
 0
 0
 0
 0
 14:03 
 0
 0
 0
 1
 12:41 
 1
 0
 0
 0
 6:21 
 0
 0
 0
 0
 15:14 
 0
 0
 0
 0
 13:36 
 0
 0
 0
 0
 12:50 
 0
 0
 0
 0
 5:04 
 1
 0
 0
 0
 13:50 
 0
 0
 0
 0
 5:02 
 0
 0
 0
 0
 2:57 
 0
 0
 0
 0
 11:50 
 0
 0
 0
 0
 5:21 
 0
 0
 0
 0
 15:37 
 0
 0
 0
 0
 3:05 
 0
 0
 1
 1
 14:45 
 0
 0
 0
 0
 15:12 
 0
 0
 0
 0
 12:55 
 0
 0
 0
 0
 14:30 
 0
 0
 0
 0
 8:28 
 0
 0
 0
 0
 15:13 
 0
 0
 0
 0
 4:55 
 0
 0
 0
 0
 3:05 
 0
 0
 0
 0
 12:13 
 0
 0
 0
 1
 5:54 
 0
 0
 0
 0
 3:17 
 0
 0
 0
 0
 15:04 
 0
 0
 0
 0
 12:23 
 0
 0
 0
 0
 21:08 
 0
 0
 0
 0
 4:23