Talks

Research Presentations

• Local Optimization Achieves Global Optimality in Multi-Agent Reinforcement Learning
  [Slides]

• Blessing of Class Diversity in Pre-training
  [Slides]

• Optimizing the Performative Risk under Weak Convexity Assumptions
  [Slides]

• Provably Efficient Policy Optimization for Two-Player Zero-Sum Markov Games
  [Slides]

Other Talks

• Training Diffusion Models with Reinforcement Learning
  [Slides]

• Risk Sensitive Reinforcement Learning with Low-rank MDPs
  [Slides]

• Eigenface++: Face Recognition with Deep PCA
  [Slides] [Report]

• What Can Transformers Learn In-Context?
  [Slides]