Jump to Content
Alekh Agarwal

Alekh Agarwal

Alekh is a researcher in the Learning Theory team at Google, where he works on a range of questions in theoretical machine learning. Bulk of his research focuses on the theory and applications of reinforcement learning, as well as other forms of interactive learning such as bandits, active learning and online learning.
Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, desc
  • Year
  • Year, desc
    Preview abstract We propose a general framework to design posterior sampling methods for model-based RL. We show that the proposed algorithms can be analyzed by reducing regret to Hellinger distance in conditional probability estimation. We further show that optimistic posterior sampling can control this Hellinger distance, when we measure model error via data likelihood. This technique allows us to design and analyze unified posterior sampling algorithms with state-of-the-art sample complexity guarantees for many model-based RL settings. We illustrate our general result in many special cases, demonstrating the versatility of our framework. View details
    Preview abstract POMDPs capture a broad class of decision making problems, but hardness results suggest that learning is intractable even in simple settings due to the inherent partial observability. However, in many realistic problems, more information is either revealed or can be computed during some point of the learning process. Motivated by diverse applications ranging from robotics to data center scheduling, we formulate a Hindsight Observable Markov Decision Process (HOMDP) as a POMDP where the latent states are revealed to the learner in hindsight and only during training. We introduce new algorithms for the tabular and function approximation settings that are provably sample-efficient with hindsight observability, even in POMDPs that would otherwise be statistically intractable. We give a lower bound showing that the tabular algorithm is optimal in its dependence on latent state and observation cardinalities. View details
    No Results Found