Pinned Loading

1. hybridcov hybridcov Public

   Source code for a paper on how optimism can boost sample efficiency within hybrid RL, accepted to RLC 2024.

   Jupyter Notebook 1 1

2. mdpmix mdpmix Public

   Source code for learning mixtures of Markov chains and MDPs, accepted as a short oral presentation to ICML 2023.

   Jupyter Notebook 1

3. off-policy off-policy Public

   Source code for off-policy evaluation and optimization in the presence of unobserved confounding, accepted to AISTATS 2024.

   Jupyter Notebook 1

4. diffPomp diffPomp Public

   Estimation, filtering, and inference for partially-observed Markov processes via a two-stage algorithm involving gradient descent (with a novel gradient estimate for the particle filter) warm-start…

   Jupyter Notebook 2 1

5. probono probono Public

   Algorithms for accelerating personalization (in the context of linear bandit recommendations) to new users, given access to embeddings from other users with heterogenous tastes and preferences. Nov…

   Jupyter Notebook 1