Toward Efficient Multi-Agent Exploration With Trajectory Entropy Maximization

Authors: Tianxu Li, Kun Zhu

ICLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental To validate the effectiveness of our method, we test our method in challenging multi-agent tasks from several MARL benchmarks. The results demonstrate that our method consistently outperforms existing state-of-the-art methods. In this section, we examine the performance of our proposed TEE method using challenging multi-agent tasks from Pac-Men, SMAC, and SMACv2 benchmarks, demonstrating its superior effectiveness.
Researcher Affiliation Academia 1College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, China 2Collaborative Innovation Center of Novel Software Technology and Industrialization EMAIL
Pseudocode Yes For the Py Torch-style pseudocode of TEE, please refer to Appendix E. The source code of our method can be found in the supplemental material.
Open Source Code Yes For the Py Torch-style pseudocode of TEE, please refer to Appendix E. The source code of our method can be found in the supplemental material.
Open Datasets Yes In this section, we examine the performance of our proposed TEE method using challenging multi-agent tasks from Pac-Men, SMAC, and SMACv2 benchmarks... The StarCraft Multi-Agent Challenge (SMAC) (Samvelyan et al., 2019)... SMACv2 benchmark (Ellis et al., 2022).
Dataset Splits No The paper describes the experimental environments and training/testing procedures (e.g., 'We set the evaluation interval to 10K steps followed by 32 test episodes. We run all methods for 5 million steps.'), but does not provide traditional training/validation/test dataset splits as commonly found in supervised learning tasks since it uses interactive reinforcement learning environments.
Hardware Specification Yes We implemented our method using Num Py and Py Torch, and all experiments are conducted on a single NVIDIA Ge Force RTX 4090 GPU.
Software Dependencies Yes The SC2.4.10 version of Star Craft II is used, and it s important to note that performance comparisons between different versions are not applicable.
Experiment Setup Yes The hyperparameters for TEE and baseline methods in Pac-Men, SMAC, and SMACv2 are detailed in Table 2.