IFML
We are the National AI Institute for Foundations of Machine Learning (IFML)
Designated by the National Science Foundation (NSF) in 2020, IFML develops the key foundational tools for the next decade of AI innovation. Our research focuses on core foundational challenges integrating mathematical tools with real-world objectives to advance the state-of-the-art.
Семинар IFML: 21.11.25 — Динамика обучения в многопользовательских играх
Lecture 25: Dheeraj Nagaraj (guest lecture): Interleaved Gibbs Diffusion
Lecture 24: Sanjay Shakkottai: ELBO for masked diffusion and discrete flow models
Lecture 23: Sanjay Shakkottai: Diffusion language models
Семинар IFML: 14.11.25 — Модели языка более быстрого распространения
Lecture 22: Advait Parulekar (guest lecture): Annealed posterior sampling
Основной доклад исследования: Вахаб Миррокни
Основной докладчик исследования: Амин Карбаси
Понимание мультимодальной активности — Кристен Грауман
Deep Proteins Research: Danny Diaz
Diffusion Models Research: Sanjay Shakkottai
Исследовательский симпозиум GenAI — осень 2025 г. (Введение — Адам Кливанс)
IFML Seminar: 11/07/25 - Model Self-improvement via Optimal Retraining
Lecture 21: Advait Parulekar (guest lecture): Hardness of posterior sampling
AIHealthTalk: 06.11.25 — Семантика в медицине: экспертные взгляды, данные и применение
AIHealthTalk: 09/11/25 - Clinical Deployment of AI:From Single Models to Compound Agentic Systems
AIHealthTalk: 25.09.2025 — PanEcho: на пути к полной интерпретации эхокардиографии с использовани...
AIHealthTalk:10/9/25 - Using Large Language Models to Simulate Patients for Training Mental Health
AIHealthTalk: 10.03.24 с Тяньлуном Ченом
AIHealthTalk: 03.04.24 с Пранавом Раджпуркаром, доктором наук
AIHealthTalk: 3/27/24 - Shaping the creation and adoption of large language models in healthcare
Lecture 20: Sanjay Shakkottai: Flow matching and Image Editing with Flow Models
AIHealth Talk: 23.10.25 — Прогнозирование долгосрочной смертности при ХОБЛ с использованием марке...
Lecture 19: Sanjay Shakkottai: Image Editing and Stylization, and Intro to Flow Matching
Lecture 18: Litu Rout (guest lecture): Inverse problems in Image Generation and Editing
Семинар IFML: 24.10.25 — Обучение по многим траекториям
Lecture 17: Sanjay Shakkottai: Posterior Sampling
Lecture 16: Sanjay Shakkottai: Rectified Flow and Intro to Posterior Sampling
IFML Seminar: 10/17/25 - Sample-Efficient Personalized Reward Models for Pluralistic Alignment
Lecture 15: Sanjay Shakkottai: Information Theoretic Approach for DDPM Discretization