Playing 2048 With Reinforcement Learning

TL;DR

This paper explores reinforcement learning methods, including deep Q-learning and beam search, to improve success rates in the game 2048, achieving a win rate of 28.5%.

Contribution

It introduces reinforcement learning techniques, particularly deep Q-learning and beam search, applied to 2048, demonstrating significant improvements in winning frequency.

Findings

Beam search achieved a 28.5% win rate in 2048.

Deep Q-learning was used as part of the approach.

The methods outperform random play significantly.

Abstract

The game of 2048 is a highly addictive game. It is easy to learn the game, but hard to master as the created game revealed that only about 1% games out of hundreds million ever played have been won. In this paper, we would like to explore reinforcement learning techniques to win 2048. The approaches we have took include deep Q-learning and beam search, with beam search reaching 2048 28.5 of time.