A Study on Overfitting in Deep Reinforcement Learning

by   Chiyuan Zhang, et al.

Recent years have witnessed significant progresses in deep Reinforcement Learning (RL). Empowered with large scale neural networks, carefully designed architectures, novel training algorithms and massively parallel computing devices, researchers are able to attack many challenging RL problems. However, in machine learning, more training power comes with a potential risk of more overfitting. As deep RL techniques are being applied to critical problems such as healthcare and finance, it is important to understand the generalization behaviors of the trained agents. In this paper, we conduct a systematic study of standard RL agents and find that they could overfit in various ways. Moreover, overfitting could happen "robustly": commonly used techniques in RL that add stochasticity do not necessarily prevent or detect overfitting. In particular, the same agents and learning algorithms could have drastically different test performance, even when all of them achieve optimal rewards during training. The observations call for more principled and careful evaluation protocols in RL. We conclude with a general discussion on overfitting in RL and a study of the generalization behaviors from the perspective of inductive bias.


page 7

page 13

page 14

page 16

page 17

page 18


A Dissection of Overfitting and Generalization in Continuous Reinforcement Learning

The risks and perils of overfitting in machine learning are well known. ...

Local Feature Swapping for Generalization in Reinforcement Learning

Over the past few years, the acceleration of computing resources and res...

The Primacy Bias in Deep Reinforcement Learning

This work identifies a common flaw of deep reinforcement learning (RL) a...

Group Equivariant Deep Reinforcement Learning

In Reinforcement Learning (RL), Convolutional Neural Networks(CNNs) have...

How to Make Deep RL Work in Practice

In recent years, challenging control problems became solvable with deep ...

An Empirical Study on Hyperparameters and their Interdependence for RL Generalization

Recent results in Reinforcement Learning (RL) have shown that agents wit...

Efficient Deep Reinforcement Learning Requires Regulating Overfitting

Deep reinforcement learning algorithms that learn policies by trial-and-...

Please sign up or login with your details

Forgot password? Click here to reset