BooVI: Provably Efficient Bootstrapped Value Iteration

Authors: Boyi Liu, Qi Cai, Zhuoran Yang, Zhaoran Wang

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical While this paper is mainly on algorithmic and theoretical aspects of the bootstrapping idea in RL, it would be interesting to see the empirical strength of Boo VI in more challenging RL environments. We leave this part to our future work.
Researcher Affiliation Academia Northwestern University; boyiliu2018@u.northwestern.edu Northwestern University; qicai2022@u.northwestern.edu Princeton University; zy6@princeton.edu Northwestern University; zhaoranwang@gmail.com
Pseudocode Yes Algorithm 1 Bootstrapped Value Iteration (Boo VI) ... Algorithm 2 Bootstrapping Action-Value Function
Open Source Code No The paper does not provide any specific repository links or explicit statements about code availability. It mentions: 'We leave this part to our future work.'
Open Datasets No The paper is primarily theoretical and does not report using a specific publicly available dataset for empirical evaluation. It refers to illustrative experiments in Appendix F, which is not provided, and states 'We leave this part to our future work' regarding empirical strength.
Dataset Splits No The paper is primarily theoretical and does not provide specific dataset split information.
Hardware Specification No The paper is primarily theoretical and does not provide specific hardware details used for running experiments.
Software Dependencies No The paper is primarily theoretical and does not provide specific software names with version numbers.
Experiment Setup No The paper is primarily theoretical and does not contain specific experimental setup details like hyperparameter values or training configurations.