# Health-Informed Policy Gradients for Multi-Agent Reinforcement Learning

arXiv: https://arxiv.org/abs/1908.01022 · HTML: https://mlsanity.com/paper/1908.01022/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Aug 2019.

## Built on by (most cited first, 2 of 2)

- [Any-Play: An Intrinsic Augmentation for Zero-Shot Coordination](https://mlsanity.com/paper/2201.12436/) (2022)
- [On Multi-Agent Deep Deterministic Policy Gradients and their Explainability for SMARTS Environment](https://mlsanity.com/paper/2301.09420/) (2023)

## Abstract (arXiv)

This paper proposes a definition of system health in the context of multiple agents optimizing a joint reward function. We use this definition as a credit assignment term in a policy gradient algorithm to distinguish the contributions of individual agents to the global reward. The health-informed credit assignment is then extended to a multi-agent variant of the proximal policy optimization algorithm and demonstrated on particle and multiwalker robot environments that have characteristics such as system health, risk-taking, semi-expendable agents, continuous action spaces, and partial observability. We show significant improvement in learning performance compared to policy gradient methods that do not perform multi-agent credit assignment.