Skip to main content

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

5
  • 5
    This is amazing! I just spent hours optimizing weights for a good heuristic function for expectimax and I implement this in 3 minutes and this completely smashes it. Commented May 29, 2014 at 17:09
  • 1
    One significant improvement, with around 15 minutes of work, will be using better estimation formula, I believe. If you are using Pure MCTS, I would suggest using the UCT formula, which is beautifully simple, but expands the decision tree in much better way. Commented Mar 7, 2015 at 21:31
  • 1
    I played around with this idea some more and did a bit of analysis if anyone is interested github.com/silverstar194/2048-ai-monte-carlo Commented Jan 2, 2017 at 4:35
  • 1
    Nice work @JohnDown! It would be nice of you if you referenced my work. Commented Jan 3, 2017 at 7:17
  • Excellent. What actually determines the best score at each node? Is it number of tile (similar value) joined? Commented Aug 9, 2022 at 20:37