# Joint action loss for proximal policy optimization

arXiv: https://arxiv.org/abs/2301.10919 · HTML: https://mlsanity.com/paper/2301.10919/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 11 references that are in the map.
- First posted on arXiv 26 Jan 2023.

## Builds on (most cited first, 8 of 8)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [Dota 2 with Large Scale Deep Reinforcement Learning](https://mlsanity.com/paper/1912.06680/) (2019)
- [Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO](https://mlsanity.com/paper/2005.12729/) (2020)
- [Mastering Complex Control in MOBA Games with Deep Reinforcement Learning](https://mlsanity.com/paper/1912.09729/) (2019)
- [Truly Proximal Policy Optimization](https://mlsanity.com/paper/1903.07940/) (2019)
- [Trust Region-Guided Proximal Policy Optimization](https://mlsanity.com/paper/1901.10314/) (2019)
- [Gym-μRTS: Toward Affordable Full Game Real-time Strategy Games Research with Deep Reinforcement Learning](https://mlsanity.com/paper/2105.13807/) (2021)

## Abstract (arXiv)

PPO (Proximal Policy Optimization) is a state-of-the-art policy gradient algorithm that has been successfully applied to complex computer games such as Dota 2 and Honor of Kings. In these environments, an agent makes compound actions consisting of multiple sub-actions. PPO uses clipping to restrict policy updates. Although clipping is simple and effective, it is not efficient in its sample use. For compound actions, most PPO implementations consider the joint probability (density) of sub-actions, which means that if the ratio of a sample (state compound-action pair) exceeds the range, the gradient the sample produces is zero. Instead, for each sub-action we calculate the loss separately, which is less prone to clipping during updates thereby making better use of samples. Further, we propose a multi-action mixed loss that combines joint and separate probabilities. We perform experiments in Gym-μRTS and MuJoCo. Our hybrid model improves performance by more than 50% in different MuJoCo environments compared to OpenAI's PPO benchmark results. And in Gym-μRTS, we find the sub-action loss outperforms the standard PPO approach, especially when the clip range is large. Our findings suggest this method can better balance the use-efficiency and quality of samples.