Multi-Expert Distillation for Few-Shot Coordination (Student Abstract)

Abstract

Ad hoc teamwork is a crucial challenge that aims to design an agent capable of effective collaboration with teammates employing diverse strategies without prior coordination. However, current Population-Based Training (PBT) approaches train the ad hoc agent through interaction with diverse teammates from scratch, which suffer from low efficiency. We introduce Multi-Expert Distillation (MED), a novel approach that directly distills diverse strategies through modeling across-episodic sequences. Experiments show that our algorithm achieves more efficient and stable training and has the ability to improve its behavior using historical contexts. Our code is available at https://github.com/LAMDA-RL/MED.

Cite

Text

Zhu et al. "Multi-Expert Distillation for Few-Shot Coordination (Student Abstract)." AAAI Conference on Artificial Intelligence, 2024. doi:10.1609/AAAI.V38I21.30539

Markdown

[Zhu et al. "Multi-Expert Distillation for Few-Shot Coordination (Student Abstract)." AAAI Conference on Artificial Intelligence, 2024.](https://mlanthology.org/aaai/2024/zhu2024aaai-multi/) doi:10.1609/AAAI.V38I21.30539

BibTeX

@inproceedings{zhu2024aaai-multi,
  title     = {{Multi-Expert Distillation for Few-Shot Coordination (Student Abstract)}},
  author    = {Zhu, Yujian and Ding, Hao and Zhang, Zongzhang},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2024},
  pages     = {23717-23719},
  doi       = {10.1609/AAAI.V38I21.30539},
  url       = {https://mlanthology.org/aaai/2024/zhu2024aaai-multi/}
}