Fast active learning for pure exploration in reinforcement learning

Authors: Pierre Menard, Omar Darwiche Domingues, Anders Jonsson, Emilie Kaufmann, Edouard Leurent, Michal Valko

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

Reproducibility Variable Result LLM Response
Research Type Experimental Empirically, we illustrate how 1/n bonuses can be beneficial for exploration in Appendix G.
Researcher Affiliation Collaboration 1Otto von Guericke University 2Inria 3Université de Lille 4Universitat Pompeu Fabra 5Deep Mind Paris.
Pseudocode Yes Algorithm 1 RF-Express
Open Source Code No "rlberry A Reinforcement Learning Library for Research and Education. https://github.com/rlberry-py/rlberry, 2021a." This is a general library, not explicitly stated to be the source code for the methodology in this specific paper.
Open Datasets No No, the paper is theoretical and focuses on algorithm design and sample complexity bounds for reinforcement learning in MDPs. It does not describe experiments using a specific dataset that would require access information for training.
Dataset Splits No No, the paper is theoretical and does not describe experiments that involve dataset splits for validation. The focus is on theoretical bounds and algorithm design.
Hardware Specification No No, the paper is theoretical and does not describe hardware used for experiments. The provided text focuses on algorithmic design and theoretical analysis.
Software Dependencies No No, the paper is theoretical and focuses on mathematical proofs and algorithms. It does not specify software dependencies with version numbers for experimental reproducibility.
Experiment Setup No No, the paper is theoretical and does not describe specific experimental setups, hyperparameters, or training configurations. The content focuses on algorithm descriptions and theoretical properties.