36-712 · Fall 2026

Hypothesis Testing: A Minimax Perspective

Overview

Description

Hypothesis testing is a fundamental task in statistics: given data, we want to decide which of two competing explanations, or hypotheses, generated it. As statistical models become higher-dimensional and more complex, determining whether reliable testing is possible—and designing procedures that achieve optimal performance—becomes increasingly challenging. This course will introduce minimax hypothesis testing, which aims to characterize the smallest separation between hypotheses that permits reliable testing. Our primary focus will be on the interplay between optimal tests and matching information-theoretic lower bounds, as well as on understanding when testing is easier than learning the underlying distribution.

Logistics

Instructor: Ankit Pensia
Email: ankitp@cmu.edu
Course number: 36-712 (Fall 2026)
Lectures: Monday and Wednesday, 9:30 AM–10:50 AM, Wean Hall 5316
Prerequisites: Graduate probability, mathematical statistics, linear algebra, and basic concentration inequalities

Schedule

The schedule will be updated as the course progresses. Topics and readings are tentative.

Hypothesis Testing: A Minimax Perspective course schedule and readings
Date Topic Reading
October 19 Introduction

References

Topics will be selected primarily from the following references. Additional references will be introduced as needed.