PAC Reinforcement Learning for Predictive State Representations

Authors: Wenhao Zhan, Masatoshi Uehara, Wen Sun, Jason D. Lee

ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our algorithm naturally works with function approximation to extend to systems with potentially large state and observation spaces. We show that given a realizable model class, the sample complexity of learning the near optimal policy only scales polynomially with respect to the statistical complexity of the model class, without any explicit polynomial dependence on the size of the state and observation spaces. Notably, our work is the first work that shows polynomial sample complexities to compete with the globally optimal policy in PSRs.
Researcher Affiliation Academia Anonymous authors Paper under double-blind review
Pseudocode Yes Algorithm 1 CRANE
Open Source Code No No, the paper does not provide any statement or link indicating the availability of open-source code.
Open Datasets No No, the paper is theoretical and focuses on sample complexity. It does not describe experiments using specific datasets, thus no access information for training data is provided.
Dataset Splits No No, the paper is theoretical and does not describe experiments with dataset splits, thus no validation split information is provided.
Hardware Specification No No, the paper is theoretical and does not mention any hardware specifications for running experiments.
Software Dependencies No No, the paper is theoretical and does not mention any specific software dependencies with version numbers.
Experiment Setup No No, the paper is theoretical and does not describe an experimental setup with hyperparameters or training settings.