Robustness in High-Dimensional Inference Tasks

on YouTube ↗43m

" summary

Claims a practical test for a single linear coefficient that works when the regression vector is dense, using an IV-style rewrite of the null and a constrained debiased estimator that enforces bounded residuals and variance estimation. Proves a new minimax detection boundary via the concept of uniform non-testability, shows the method controls Type I error when p is exponentially larger than n and outperforms standard debiasing in simulations and in breast‑cancer and macroeconomic examples.

AI-written summary · about

" chapters (13)

1. 0:00Introduction
2. 2:05Setting
3. 4:55Plot
4. 6:45Literature Review
5. 14:44Moment Condition
6. 15:34Constraint Dancing
7. 20:47Linear Contrast
8. 26:29Conditions
9. 28:24Linear Model
10. 29:01Robustness Property
11. 31:53Uniform NonTestability
12. 36:18Numerical Experiments
13. 37:48Plots

" related talks

• 57mA New Perspective on High-Dimensional Causal InferencePragya Sur · Deep Learning Theory Workshop and Summer School 2022
• 22mPanel DiscussionChristopher Harshaw, Liam Solus, Arnab Bhattacharyya · Learning from Interventions 2022
• 45mHigh Dimensional Robust Sparse RegressionConstantine Caramanis · Sublinear Algorithms and Nearest-Neighbor Search
• 01h 07mRecent Progress in High-Dimensional LearningAnkur Moitra · Probability, Geometry, and Computation in High Dimensions Boot Camp (Simons Institute, Fall 2020)
• 53mAdversarial Machine Learning and Instrumental Variables for Flexible Causal ModelingVasilis Syrgkanis · Adversarial Approaches in Machine Learning 2022
• 45mThe Statistical Fairness-Accuracy FrontierAlireza Fallah · Learning from Heterogeneous Sources 2026
• 34mDistributional Robustness, Learning, and Empirical LikelihoodJohn Duchi · Optimization, Statistics and Uncertainty
• 01h 30mHigh-Dimensional Statistics IMartin Wainwright · Big Data Boot Camp 2013