REVIEW 2 cited by
Provably Efficient Reinforcement Learning with Aggregated States
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
We establish that an optimistic variant of Q-learning applied to a fixed-horizon episodic Markov decision process with an aggregated state representation incurs regret $\tilde{\mathcal{O}}(\sqrt{H^5 M K} + \epsilon HK)$, where $H$ is the horizon, $M$ is the number of aggregate states, $K$ is the number of episodes, and $\epsilon$ is the largest difference between any pair of optimal state-action values associated with a common aggregate state. Notably, this regret bound does not depend on the number of states or actions and indicates that asymptotic per-period regret is no greater than $\epsilon$, independent of horizon. To our knowledge, this is the first such result that applies to reinforcement learning with nontrivial value function approximation without any restrictions on transition probabilities.
Forward citations
Cited by 2 Pith papers
-
Commit to the Bit: Reactive Reinforcement Learning Done Right
Committed Q-learning converges almost surely to the optimal reactive policy under the rewire-robustness assumption, which is strictly weaker than q*-realizability.
-
Concurrent Learning with Aggregated States via Randomized Least Squares Value Iteration
Concurrent RLSVI with aggregated states is shown to have worst-case regret O~(K H^(5/2) Γ √N) and per-agent regret 1/√N, with an analogous infinite-horizon bound.
Discussion (0). Continue with ORCID to comment.