Conference: 2026 Princeton Machine Learning Theory Summer School

This award supports 2026 Princeton Machine-Learning Theory Summer School that will be held at Princeton University on August 3 - August 14, 2026. The school will host 80 - 90 PhD students from around the world and will consist of six courses given by top scholars. This series started in 2021 with a core goal to form a cohesive community of machine learning theory researchers across math, statistics, physics, computer science, and electrical engineering. This is important for US work force development and overall competitiveness in critical national security areas of AI / ML. Moreover, through speakers (e.g. from Meta, IBM, Google, Salesforce) and sponsors (e.g. DARPA, Jane Street) these summer schools open a significant surface area for interaction between industry, government, and academia. Modern machine learning, notably the use of neural network-based algorithms, has led to significant progress in domains from natural language processing (e.g. ChatGPT) to structural biology (e.g. AlphaFold) and computer vision (e.g. self-driving cars). Most breakthroughs underpinning deep learning are the result of a brilliant engineering based on theoretically-grounded principles. The latter include principled optimizer design, quantization schemes, architecture selection (e.g. diffusion models), and model scaling strategies (e.g. scaling laws or hyperparameter transfer). Behind any such principled approach to machine learning is a community of students and researchers at the intersection of mathematics, physics, statistics, and computer science. Entering this community is often difficult for students because the field is still intellectually fragmented, with few systematic educational resources (e.g. textbooks, summer schools, workshops) and opportunities to meet other students and junior researchers interested in the subject. Princeton Machine-Learning Theory Summer Schools is one of the ways to address these challenges. The meeting website is at https://mlschool.princeton.edu/ This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2620782 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Boris Hanin | Institution: Princeton University, PRINCETON, NJ | Award Amount: $49,999 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2620782 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2620782.html