Monte-Carlo Tree Search in Continuous Action Spaces with Value Gradients

Authors: Jongmin Lee, Wonseok Jeon, Geon-Hyeong Kim, Kee-Eung Kim4561-4568

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In the experiments, we demonstrate that our approach outperforms existing MCTS methods and other strong baseline algorithms for continuous action spaces.
Researcher Affiliation Academia 1School of Computing, KAIST, Daejeon, Republic of Korea 2Graduate School of AI, KAIST, Daejeon, Republic of Korea 3Mila, Quebec AI Institute, Montreal, Canada 4Mc Gill University, Montreal, Canada
Pseudocode Yes Algorithm 1 Value-Gradient UCT (VG-UCT)
Open Source Code No The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper.
Open Datasets Yes Realistic experiments were conducted on four stochastic variants of continuous control tasks: Pendulum (continuous action version), Acrobot (continuous action version), Reacher and Pusher from the Open AI Gym environment (Brockman et al. 2016; Todorov, Erez, and Tassa 2012)
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper mentions using the "Open AI Gym environment" but does not specify version numbers for Gym or other ancillary software dependencies, which are necessary for replication.
Experiment Setup No Detailed experimental settings are provided in Appendix D.