Paper-Conference

On Diffusion Models for Multi-Agent Partial Observability: Shared Attractors, Error Bounds, and Composite Flow

We show that diffusion models can reconstruct global states in decentralized partially observable multiagent systems, with approximation errors leading to deviations that can be bounded for convergence to the true state.

Tonghan Wang, Heng Dong (董恒), Yanchen Jiang, David C. Parkes, Milind Tambe

On Diffusion Models for Multi-Agent Partial Observability: Shared Attractors, Error Bounds, and Composite Flow

Enhancing Decision-Making of Large Language Models via Actor-Critic

We propose an LLM-based Actor-Critic algorithm that integrates actor and critic methods in the way that would utilize the merits of the actor-critic algorithm with the strengths of LLMs.

Heng Dong (董恒), Kefei Duan, Chongjie Zhang

Enhancing Decision-Making of Large Language Models via Actor-Critic

Leveraging Hyperbolic Embeddings for Coarse-to-Fine Robot Design

We propose to design multi-cellular robots in a coarse-to-fine manner and leverage hyperbolic embeddings for realization.

Heng Dong (董恒), Junyu Zhang, Chongjie Zhang

Leveraging Hyperbolic Embeddings for Coarse-to-Fine Robot Design

Symmetry-Aware Robot Design with Structured Subgroups

We exploit the structure of the design space in robot design problems with symmetry characteristics and generate robots with high performance more efficiently.

Heng Dong (董恒), Junyu Zhang, Tonghan Wang, Chongjie Zhang

Symmetry-Aware Robot Design with Structured Subgroups

Low-Rank Modular Reinforcement Learning via Muscle Synergy

Synergy-Oriented LeARning Framework (SOLAR).

Heng Dong (董恒), Tonghan Wang, Jiayuan Liu, Chongjie Zhang

Low-Rank Modular Reinforcement Learning via Muscle Synergy

Birds of a feather flock together: A close look at cooperation emergence via multi-agent rl

Use the idea of homophily to solve second-order social dilemmas.

Heng Dong (董恒), Tonghan Wang, Jiayuan Liu, Chi Han, Chongjie Zhang

DOP: Off-Policy Multi-Agent Decomposed Policy Gradients

Multi-agent decomposed policy gradient.

Yihan Wang, Beining Han, Tonghan Wang, Heng Dong (董恒), Chongjie Zhang

DOP: Off-Policy Multi-Agent Decomposed Policy Gradients

ROMA: Multi-Agent Reinforcement Learning with Emergent Roles

Role-oriented MARL.

Tonghan Wang, Heng Dong (董恒), Victor Lesser, Chongjie Zhang

ROMA: Multi-Agent Reinforcement Learning with Emergent Roles