Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO

05/25/2020
by   Logan Engstrom, et al.
0

We study the roots of algorithmic progress in deep policy gradient algorithms through a case study on two popular algorithms: Proximal Policy Optimization (PPO) and Trust Region Policy Optimization (TRPO). Specifically, we investigate the consequences of "code-level optimizations:" algorithm augmentations found only in implementations or described as auxiliary details to the core algorithm. Seemingly of secondary importance, such optimizations turn out to have a major impact on agent behavior. Our results show that they (a) are responsible for most of PPO's gain in cumulative reward over TRPO, and (b) fundamentally change how RL methods function. These insights show the difficulty and importance of attributing performance gains in deep reinforcement learning. Code for reproducing our results is available at https://github.com/MadryLab/implementation-matters .

READ FULL TEXT
research
01/22/2022

Bag of Tricks for Natural Policy Gradient Reinforcement Learning

Natural policy gradient methods are popular reinforcement learning metho...
research
07/05/2021

Ensemble and Auxiliary Tasks for Data-Efficient Deep Reinforcement Learning

Ensemble and auxiliary tasks are both well known to improve the performa...
research
03/22/2022

Is Vanilla Policy Gradient Overlooked? Analyzing Deep Reinforcement Learning for Hanabi

In pursuit of enhanced multi-agent collaboration, we analyze several on-...
research
07/02/2018

Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization

This paper proposes a first order gradient reinforcement learning algori...
research
02/08/2021

RL-Scope: Cross-Stack Profiling for Deep Reinforcement Learning Workloads

Deep reinforcement learning (RL) has made groundbreaking advancements in...
research
08/28/2020

Next-Best View Policy for 3D Reconstruction

Manually selecting viewpoints or using commonly available flight planner...

Please sign up or login with your details

Forgot password? Click here to reset