Sign inSign in

Simons Institute (UC Berkeley)

Research Institute · Berkeley, California · founded 2012

The Simons Institute for the Theory of Computing is a research institute at UC Berkeley in Berkeley, California, founded in 2012 to hold collaborative research programs.

More

The Simons Institute for the Theory of Computing is a research institute at UC Berkeley in Berkeley, California, founded in 2012 to hold collaborative research programs. Its lectures here are graduate-level public colloquia in AI and machine learning, standing apart from any course.

lectures
63
courses
2
talks
20
scholars
64
hours
43
The Conscious Turing Machine (CTM), a formally defined Theoretical model of Consciousness45:00
Decision Theoretic Foundations for Human-AI Collaboration31:01
AI and Safety29:26
Scheduling with Time-Evolving Uncertainty for Content Review in Social Media49:19
The Inadequacy of Offline LLM Evaluations: Personalization and Model Behavior31:57
Theory of Modern AI: Learning Theoretic, Game Theoretic, and Algorithmic Perspectives40:41
Venturing out into the world with AI44:41
Astral Space: Convex Analysis at Infinity44:01
Agreement and Alignment for Human-AI Collaborative Decision Making45:21
Multi-objective Learning: An Algorithmic Toolbox for Optimal Predictions Anytime Anywhere37:40
Red Points and Blue Points38:15
Probably Approximately Precision and Recall Learning34:39
Efficient Algorithms for Reliable Machine Learning37:26
The Role of Standardized Tests in Admissions: Predictions vs. Decisions33:37
Recommendation in Equilibrium for the NYC High School Match32:51
Enhanced and Efficient Reasoning in Large Language Models36:16
An Occam's Razor Principle for Transformers?39:55
The Feedback Loop of Statistical Discrimination32:03
Computational Thinking on Learning Models38:30
LLM-Guided Reinforcement Learning for Mastery Learning: Large-Scale Field Evidence35:06
On the Komlós Conjecture33:41
Human Values in Clinical Algorithms, Old and New38:46
Approximately Packing Dijoins Via Nowhere-Zero Flows33:46
Individual Gaps in Prediction Generalization30:41
Evaluation Design29:16
Theoretical Foundations of Modern Machine Learning Paradigms38:15
Questions for Theory in the New Age of Machine Learning42:01
Non-parametric Causal Inference in Dynamic Thresholding Designs30:21
Foundations, Privacy & Generalization Challenges in Modern Machine Learning36:41
Unfamiliar Terrain29:51
When Prediction Becomes Policy: What Do Risk Models Actually Learn?33:31
Triage Score: A Counterfactual Risk Assessment Instrument39:56
Operational Dosage: The Impact of Capacity Constraints on RCT Evaluation35:01
Latent Variable models and Subset Smoothing29:26
Algorithmic UDAP32:13
Disincentivizing Hallucination in Language Models33:40
The Challenge of Valid Evaluations32:41
The Role of TCS in Understanding (AI) Consciousness43:21