Warm-up Free Policy Optimization: Improved Regret in Linear Markov Decision Processes

Authors: Asaf Cassel, Aviv Rosenberg

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical This paper eliminates this undesired warm-up phase, replacing it with a simple and efficient contraction mechanism. Our PO algorithm achieves rate-optimal regret with improved dependence on the other parameters of the problem (horizon and function approximation dimension) in two fundamental settings: adversarial losses with full-information feedback and stochastic losses with bandit feedback.
Researcher Affiliation Collaboration Asaf Cassel Tel Aviv University acassel@mail.tau.ac.il Aviv Rosenberg Google Research avivros@google.com
Pseudocode Yes Algorithm 1 Contracted Features PO for linear MDPs
Open Source Code No The paper does not contain any explicit statement about releasing open-source code for the described methodology, nor does it provide a link to a code repository.
Open Datasets No This paper focuses on theoretical contributions and does not involve empirical studies with specific datasets.
Dataset Splits No This paper is theoretical and does not describe any experimental validation process involving dataset splits.
Hardware Specification No The paper is theoretical and does not describe any experimental setup or the specific hardware used to run experiments.
Software Dependencies No The paper is theoretical and does not mention any specific software dependencies with version numbers for experimental reproducibility.
Experiment Setup No The paper is theoretical and does not detail an empirical experimental setup with hyperparameters or training configurations.