Viewing profile — ovolve
ovolve
HN member- Joined
- Thu, Mar 22, 2012, 5:08 PM UTC
- HN karma
- 3
- Public activity
- 10 items
- HN profile
- View on Hacker News ↗
About ovolve
No profile information was provided.
Recent public activity
-
comment
Comment #41940767
Someone forked 2048-AI and added evil mode :) https://sztupy.hu/2048-Hard/ Originating in the StackOverflow comments here https://stackoverflow.com/questions/22342854/what-is-the-o…
-
comment
Comment #7382940
Ahh good catch. That got messed up during refactoring the original code. Thanks to the person who submitted the fix on github too.
-
comment
Comment #7381414
I've been messing around trying to find a balance between both of those heuristics. They're both implemented, but there doesn't seem to be any magic bullet.
-
comment
Comment #7381400
=) Check out the eval function, and specifically the function smoothness() in grid.js. It implements the edge weighting you describe!
-
comment
Comment #7380355
Heh. If you look in the code you'll see a big commented out chunk where I tried randomly sampling computer moves to get sort of an 'expected value' for the opposition's move. Empir…
-
comment
Comment #7380310
Trying every possibility was way too slow (branching factor of ~15 to 20), so it only searches the most "annoying" moves, where annoying is defined by lining up with the highest va…
-
comment
Comment #7380275
Random location (uniform). 90% chance of a 2, 10% chance of a 4. This actually made it hard to model the computer move in the search.
-
comment
Comment #7380084
Sorry, the code is a little unkempt. The basic idea is minimax search. Googling that will get you started, but basically the algorithm plays out the game and keeps a score of the p…
-
comment
Comment #7379843
Source code here https://github.com/ov3y/2048-AI
- story