Action-Conditional Video Prediction using Deep Networks in Atari Games

TL;DR

This paper introduces deep neural network architectures for long-term, action-conditioned video frame prediction in Atari games, demonstrating realistic frame generation and potential for reinforcement learning applications.

Contribution

It presents novel deep neural network models capable of long-term, action-conditional video prediction in high-dimensional Atari game environments, a first in the field.

Findings

Generated visually-realistic frames for 100-step predictions

Models effectively incorporate control inputs into future frame predictions

First to evaluate long-term predictions conditioned on actions in high-dimensional videos

Abstract

Motivated by vision-based reinforcement learning (RL) problems, in particular Atari games from the recent benchmark Aracade Learning Environment (ALE), we consider spatio-temporal prediction problems where future (image-)frames are dependent on control variables or actions as well as previous frames. While not composed of natural scenes, frames in Atari games are high-dimensional in size, can involve tens of objects with one or more objects being controlled by the actions directly and many other objects being influenced indirectly, can involve entry and departure of objects, and can involve deep partial observability. We propose and evaluate two deep neural network architectures that consist of encoding, action-conditional transformation, and decoding layers based on convolutional neural networks and recurrent neural networks. Experimental results show that the proposed architectures are able to generate visually-realistic frames that are also useful for control over approximately 100-step action-conditional futures in some games. To the best of our knowledge, this paper is the first to make and evaluate long-term predictions on high-dimensional video conditioned by control inputs.