Minimum Rényi Correlation Principle: From Marginals to Joint Distribution

on YouTube ↗21m

" summary

Defines a Principle of Minimum Rényi Correlation that picks a joint P agreeing with given first/second marginals by minimizing Rényi maximal correlation, proves for real vectors the joint Gaussian minimizes Rényi correlation under fixed moments, gives a discrete recipe that reduces to a quadratic program on indicator variables, proves a necessary-and-sufficient separability condition that can be checked in linear time, and shows the estimator recovers group LASSO after empirical plug‑in.

AI-written summary · about

" chapters (9)

1. 0:00Introduction to modeling
2. 2:29Defining dependence measures
3. 3:32Rényi maximal correlation
4. 5:55Principle of minimum correlation
5. 7:44Solving the discrete case
6. 11:20Existence and theoretical bounds
7. 13:26Application to variable selection
8. 14:43Summary and conclusions
9. 16:14Q&A and future directions

" related talks

• 09mConvex Programs, Computational Phase Transitions, and Robust AlgorithmsSam Hopkins · Meet the Fellows Welcome Event
• 32mQuantitative Correlation Inequalities via Semigroup InterpolationAnindya De, Shivam Nadimpalli, Rocco A. Servedio · ITCS 2021
• 54mExact Matching Of Random Graphs With Constant CorrelationCheng Mao · Algorithmic Advances for Statistical Inference with Combinatorial Structure
• 52mNew Algorithms For The Graph Alignment ProblemMarc Lelarge · Algorithmic Advances for Statistical Inference with Combinatorial Structure
• 40mTesting Correlation of Unlabeled Random GraphsJiaming Xu · Computational Phase Transitions (Simons Institute, 2020)
• 41mReducibility and Statistical-Computational Gaps from Secret LeakageGuy Bresler · Computational Phase Transitions (Simons Institute, 2020)
• 39mThe Overlap Gap Property in Principal Submatrix RecoverySubhabrata Sen · Computational Phase Transitions (Simons Institute, 2020)
• 54mNegative Dependence and Lorentzian PolynomialsPetter Brändén · Deterministic Counting, Probability, and Zeros of Partition Functions (Simons Institute workshop, 2019)