When Existing Techniques in Linear Regression Preserve Differential Privacy

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


Duration: 31:18
379 views
3

Or Sheffet, University of Alberta
https://simons.berkeley.edu/talks/or-sheffet-2016-11-18
Learning, Algorithm Design and Beyond Worst-Case Analysis



Other Videos By Simons Institute for the Theory of Computing


2016-12-06Composition in Some Formal Models of Natural Language
2016-12-06Modelling Interconnected Systems with Decorated Corelations
2016-12-05From Linearizability to Eventual Consistency
2016-12-05Data Structures for Quasistrict Higher Categories
2016-12-05An Operadic Approach to Compositionality
2016-12-05Compositionality, Adequacy, and Full Abstraction
2016-12-05Semantics for Physicists
2016-12-01Logic and Automata and Beyond
2016-11-18The Power of Predictions in Online Optimization
2016-11-18Above Average-Case Analysis?
2016-11-18When Existing Techniques in Linear Regression Preserve Differential Privacy
2016-11-18Monotone Estimation Framework and its Applications for Scalable Analytics of Large Data Sets
2016-11-18How Hard Is Inference for Structured Prediction?
2016-11-18Automated Scientific Discovery Using Insights from Problem Structure
2016-11-17Sketching and Randomization for Distributed Submodular and Coverage Optimization
2016-11-17When Does Clustering Become Easy, and Should We Care About Other Cases?
2016-11-17Follow the Leader with Dropout Perturbations
2016-11-17Beyond Worst Case: When Complex Feedback Can Improve the Label Complexity of Active Learning
2016-11-17Characterizing the Typical Case Complexity of Formal Verification and Synthesis
2016-11-17Recovery Guarantee of Non-Negative Matrix Factorization via Alternating Updates
2016-11-17The Logic of Counting Query Answers


Tags:
Or Sheffet
Learning Algorithm Design and Beyond Worst-Case Analysis
Simons Institute
Theory of Computing
Theory of Computation
Theoretical Computer Science
Computer Science
UC Berkeley