# Phasic Policy Gradient

arXiv: https://arxiv.org/abs/2009.04416 · HTML: https://mlsanity.com/paper/2009.04416/

## Key facts

- Cited by 123 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Sep 2020.

## Built on by (most cited first, 30 of 123)

- [Mastering Diverse Domains through World Models](https://mlsanity.com/paper/2301.04104/) (2023)
- [Deep Reinforcement Learning at the Edge of the Statistical Precipice](https://mlsanity.com/paper/2108.13264/) (2021)
- [Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos](https://mlsanity.com/paper/2206.11795/) (2022)
- [A Survey of Zero-shot Generalisation in Deep Reinforcement Learning](https://mlsanity.com/paper/2111.09794/) (2021)
- [Learning to drive from a world on rails](https://mlsanity.com/paper/2105.00636/) (2021)
- [Broken Neural Scaling Laws](https://mlsanity.com/paper/2210.14891/) (2022)
- [PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNav](https://mlsanity.com/paper/2301.07302/) (2023)
- [Decoupling Value and Policy for Generalization in Reinforcement Learning](https://mlsanity.com/paper/2102.10330/) (2021)
- [Rethinking the Implementation Tricks and Monotonicity Constraint in Cooperative Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2102.03479/) (2021)
- [Scaling Scaling Laws with Board Games](https://mlsanity.com/paper/2104.03113/) (2021)
- [Muesli: Combining Improvements in Policy Optimization](https://mlsanity.com/paper/2104.06159/) (2021)
- [VRL3: A Data-Driven Framework for Visual Deep Reinforcement Learning](https://mlsanity.com/paper/2202.10324/) (2022)
- [Pre-Trained Video Generative Models as World Simulators](https://mlsanity.com/paper/2502.07825/) (2025)
- [Scaling laws for single-agent reinforcement learning](https://mlsanity.com/paper/2301.13442/) (2023)
- [History Compression via Language Models in Reinforcement Learning](https://mlsanity.com/paper/2205.12258/) (2022)
- [Query-Dependent Prompt Evaluation and Optimization with Offline Inverse RL](https://mlsanity.com/paper/2309.06553/) (2023)
- [Efficient Transformers in Reinforcement Learning using Actor-Learner Distillation](https://mlsanity.com/paper/2104.01655/) (2021)
- [Batch size-invariance for policy optimization](https://mlsanity.com/paper/2110.00641/) (2021)
- [Multi-agent Reinforcement Learning: A Comprehensive Survey](https://mlsanity.com/paper/2312.10256/) (2023)
- [Stabilizing Off-Policy Deep Reinforcement Learning from Pixels](https://mlsanity.com/paper/2207.00986/) (2022)
- [Procedural Generalization by Planning with Self-Supervised World Models](https://mlsanity.com/paper/2111.01587/) (2021)
- [Bridging Evolutionary Algorithms and Reinforcement Learning: A Comprehensive Survey on Hybrid Algorithms](https://mlsanity.com/paper/2401.11963/) (2024)
- [Hyperbolic Deep Reinforcement Learning](https://mlsanity.com/paper/2210.01542/) (2022)
- [Measuring Sample Efficiency and Generalization in Reinforcement Learning Benchmarks: NeurIPS 2020 Procgen Benchmark](https://mlsanity.com/paper/2103.15332/) (2021)
- [Cross-Trajectory Representation Learning for Zero-Shot Generalization in RL](https://mlsanity.com/paper/2106.02193/) (2021)
- [Improving Policy Optimization with Generalist-Specialist Learning](https://mlsanity.com/paper/2206.12984/) (2022)
- [Explore to Generalize in Zero-Shot RL](https://mlsanity.com/paper/2306.03072/) (2023)
- [CFlowNets: Continuous Control with Generative Flow Networks](https://mlsanity.com/paper/2303.02430/) (2023)
- [Open RL Benchmark: Comprehensive Tracked Experiments for Reinforcement Learning](https://mlsanity.com/paper/2402.03046/) (2024)
- [Improving Deep Reinforcement Learning by Reducing the Chain Effect of Value and Policy Churn](https://mlsanity.com/paper/2409.04792/) (2024)

## Abstract (arXiv)

We introduce Phasic Policy Gradient (PPG), a reinforcement learning framework which modifies traditional on-policy actor-critic methods by separating policy and value function training into distinct phases. In prior methods, one must choose between using a shared network or separate networks to represent the policy and value function. Using separate networks avoids interference between objectives, while using a shared network allows useful features to be shared. PPG is able to achieve the best of both worlds by splitting optimization into two phases, one that advances training and one that distills features. PPG also enables the value function to be more aggressively optimized with a higher level of sample reuse. Compared to PPO, we find that PPG significantly improves sample efficiency on the challenging Procgen Benchmark.