E-MAPP: Efficient Multi-Agent Reinforcement Learning with Parallel Program Guidance
Authors: Can Chang, Ni Mu, Jiajun Wu, Ling Pan, Huazhe Xu
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct extensive experiments on a series of challenging, long-horizon cooperative tasks in the Overcooked environment. Results show that E-MAPP outperforms strong baselines in terms of the completion rate, time efficiency, and zero-shot generalization ability by a large margin. |
| Researcher Affiliation | Academia | Can Chang1,2 , Ni Mu3, Jiajun Wu4, Ling Pan5, Huazhe Xu1,2 1IIIS, Tsinghua University 2Shanghai Qi Zhi Institute 3Southeast University 4Stanford University 5Mila, Université de Montréal |
| Pseudocode | Yes | The complete algorithm is shown in Appendix A.3. |
| Open Source Code | Yes | Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes] In the supplementary material |
| Open Datasets | Yes | To evaluate the proposed framework, we adapted the previous environment [50]mimicking the video game to a more challenging one Overcooked v2. More details about the environment can be found in Appendix A.7. |
| Dataset Splits | Yes | Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes] In the supplementary material |
| Hardware Specification | Yes | Did you include the total amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)? [Yes] In the supplementary material |
| Software Dependencies | No | The paper states that 'code, data, and instructions needed to reproduce the main experimental results' are in the supplementary material, but it does not explicitly provide specific version numbers for software dependencies in the main text. |
| Experiment Setup | Yes | Training details and detailed architecture descriptions can be found in Appendix A.5. Training details of the auxiliary functions are in Appendix A.6. Additionally, the paper states under ethical considerations that 'all the training details (e.g., data splits, hyperparameters, how they were chosen)' are specified in the supplementary material. |