"Moving Away from Probabilities" – Nati Srebro, Research at TTIC - YouTube
youtube.com · 11,252 words · saved by 1 readers
Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
“Moving Away from Probabilities” Nati Srebro, TTIC Originally recorded on April 3, 2026, at TTIC. In this talk, Nati Srebro explores alternative approaches to probabilistic modeling in machine learning, examining frameworks that move beyond traditional probability-based methods. He discusses implications for learning theory, optimization, and the design of models that better capture real-world uncertainty and decision-making. Timestamps: 00:00 Introduction 01:10 Talk begins 49:30 Q&A #MachineLearning #AI #LearningTheory #Optimization #Statistics #Algorithms #Research #TTIC So, okay.…
saved by
related reading
- Ilya Sutskever — We're moving from the age of scaling to the age of researchdwarkesh.com
- Andrej Karpathy — AGI is still a decade awaydwarkesh.com
- Deep Dive into LLMs like ChatGPTyoutube.com
- Sparks of AGI: early experiments with GPT-4youtube.com
- AI Is Learning at the Wrong Level of Abstraction — Matthieu Wyartyoutube.com
- An Observation on Generalizationyoutube.com
- Non_Interactive – Software & MLnonint.com
- Probabilities are not the right concept — LessWronglesswrong.com
- Inference, Diffusion, World Models, and More | YC Paper Clubyoutube.com
- There's never been a better time to go into mathyoutube.com
- Some Lessons from Adversarial Machine Learning | FAR.AIfar.ai
- Musings on typicality – Sander Dielemansander.ai