Papers › Playing 2048 With Reinforcement Learning

Playing 2048 With Reinforcement Learning

20 Oct 2021arXiv:2110.10374archive 2025-07-28

Shilun Li, Veronica Peng

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.

Paper PDF Code

Code

Shilun-Allan-Li/2048-reinforcement_learning officialpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Playing the Game of 2048 Q-Learning Reinforcement Learning Reinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Playing the Game of 2048	The Game of 2048	Beam Search	Average Score	1024	#4 of 5	Archive leaderboard	report
Playing the Game of 2048	The Game of 2048	DQN (1000 episodes)	Average Score	256	#5 of 5	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Q-Learning

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections