Some past talks

• CLIMB Seminar, UC Berkeley, April 2024: Challenges with Covariate Shift: From Prediction to Causal Inference

  ---

• Statistics and Data Science Seminar, MIT March 2023 Variational Methods in Reinforcement Learning

  ---

• On-line Reinforcement Learning Theory seminar, November 2022: Policy optimization in non-parametric settings: Optimal guarantees for kernel methods

  ---

• Simons Institute Workshop, Berkeley, November 2021 Beyond Worst-Case: Instance-Dependent Optimality in Reinforcement Learning

  ---

• International Congress of Mathematicians (ICM), Seoul, Korea, August 2014: Constrained forms of statistical minimax: Computation, communication and privacy

  ---

• Centre International de Reconcontres Mathematiques, Marseille, December 2013: Privacy and Statistical Minimax: Quantitative Trade-offs

  ---

• Simons Institute Workshop, Berkeley, September 2013 High-Dimensional Statistics, Part I

  ---

• Simons Institute Workshop, Berkeley, September 2013 High-Dimensional Statistics, Part II

  ---