Learning without Interaction Requires Separation

Channel:
United States Simons Institute for the Theory of Computing
Subscribers:
68,700
Published on ● Video Link: https://www.youtube.com/watch?v=lAAk_ZOstg8


Duration: 35:45
189 views
7

Vitaly Feldman (Google)
Privacy and the Science of Data Analysis
https://simons.berkeley.edu/talks/tbd-44



Other Videos By Simons Institute for the Theory of Computing


2019-04-12A Necessary and Sufficient Stability Notion for Adaptive Generalization
2019-04-12The Limits of Post-Selection Generalization
2019-04-12On the Power of Interactivity in Local Differential Privacy
2019-04-12Smooth Sensitivity, Revisited
2019-04-12Towards a Characterization of Differentially Private Learning
2019-04-12Groups and Individuals in Fairness and Probability
2019-04-12Is Interaction Necessary for Distributed Private Learning?
2019-04-11Private Learning with Auxiliary Public Data
2019-04-11Federated Heavy Hitters Discovery with Differential Privacy
2019-04-11Private Center Points and Learning of Halfspaces
2019-04-11Learning without Interaction Requires Separation
2019-04-11Private Meta-Selection
2019-04-11Private Boosting and Noise-tolerant Learning
2019-04-11Privacy Amplification by Sampling and Renyi Differential Privacy
2019-04-11Locally Private Statistical Inference: Communication, Samples, and Shared Randomness
2019-04-10Open Problem Session
2019-04-10The Privacy Blanket of the Shuffle Model
2019-04-10Black-box Random Differential Privacy with Sensitivity Sampling
2019-04-10Pufferfish Privacy Mechanisms for Correlated Data
2019-04-10Bandit Learning under Differential Privacy
2019-04-10Differentially Private Change-Point Detection


Tags:
Simons Institute
theoretical computer science
UC Berkeley
Vitaly Feldman
theory of computing
theory of computation
computer science
Privacy and the Science of Data Analysis