More Reading

Links to some interesting papers we didn’t have time to cover in class:

• Proving Chernoff-like concentration bounds using the stability tools we have developed in class:
  • Subgaussian Tail Bounds via Stability Arguments
  • Concentration Bounds for High Sensitivity Functions via Differential Privacy
• Other Differentially Private Estimators:
  • Generalization for Adaptively Chosen Estimators via Stable Median
  • Sublinear Time Adaptive Data Analysis
• Adaptive Data Gathering
  • Why Adaptively Collected Data have Negative Bias and How to Correct For It
• Bayesian Adaptive Data Analysis
  • Challenges in Bayesian Adaptive Data Analysis
  • Bayesian Adaptive Data Analysis Guarantees from Subgaussianity
• Hypothesis Testing
  • Max-Information, Differential Privacy, and Post-Selection Hypothesis Testing
  • Differentially Private Chi-Squared Hypothesis Testing
  • A New Class of Private Chi-Squared Tests
  • Priv’IT and Sample Efficient Identity Testing
• Approaches from Statistics (Selective Inference and Post Selection Inference)
  • Exact Post-Selection Inference with Application to the Lasso
  • Optimal Inference after Model Selection
  • Selective Inference with a Randomized Response
  • Valid Post-Selection Inference
  • Inferactive Data Analysis
• Economic Approaches
  • The Quality Preserving Database: A Computational Framework for Encouraging Collaboration, Enhancing Power and Controlling False Discovery
• Information Theoretic Views of Generalization
  • Controlling Bias in Adaptive Data Analysis via Information Theory
  • Information Theoretic Analysis of Generalization Capability of Learning Algorithms
  • Learners that Leak Little Information
  • Calibrating Noise to Variance in Adaptive Data Analysis

 

See also the website for WADAPT, a (semi)-periodic workshop on adaptive data analysis, which contains some slides and other materials.