AI Paper Atlas

Research map / Multi-agent systems and games / Cooperative multi-agent RL

MIXRTs: Toward Interpretable Multi-Agent Reinforcement Learning via Mixing Recurrent Soft Decision Trees

Zichuan Liu, Yuanyang Zhu, Zhi Wang, Yang Gao, Chunlin Chen

TPAMI 2025 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Cooperative MARL algorithms · 20 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add recurrent tree cells instead of RNN individual values (ablation)MIXRTs (mixing trees) · SMAC 8m_vs_9mtest win rate (%): 78.90 -> 84.76Improved
add recurrent tree cells instead of RNN individual values (ablation)MIXRTs (mixing trees) · SMAC 5m_vs_6mtest win rate (%): 69.17 -> 71.87Improved
add recurrent tree cells instead of RNN individual values (ablation)MIXRTs (mixing trees) · SMAC 3s5ztest win rate (%): 87.89 -> 83.98Hurt
use mixing trees instead of VDN sum mixing (ablation)RNN agents · SMAC 3s5ztest win rate (%): 82.03 -> 87.89Improved
use mixing trees instead of VDN sum mixing (ablation)RNN agents · SMAC 5m_vs_6mtest win rate (%): 73.95 -> 69.17Hurt
use mixing trees instead of QTRAN mixing (ablation)RNN agents · SMAC 3s5ztest win rate (%): 17.96 -> 87.89Improved
use mixing trees instead of QTRAN mixing (ablation)RNN agents · SMAC 5m_vs_6mtest win rate (%): 55.46 -> 69.17Improved
mask 30% of least important featuresMIXRTs · SMAC 3mtest win rate (%): 100.00 -> 48.38Hurt
mask 5% of most important featuresMIXRTs · SMAC 3mtest win rate (%): 100.00 -> 83.48Hurt
mask 5% of most important featuresMIXRTs · SMAC 5m_vs_6mtest win rate (%): 63.37 -> 13.37Hurt
mask 20% of least important featuresMIXRTs · SMAC 2s3ztest win rate (%): 100.00 -> 70.96Hurt
increase depth of individual RTCs and mixing trees (ablation)MIXRTs · SMACtest win rate (%)Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy