Shall We Play A Game
summary
Builds a complete Chomp AI that actually beats humans on a 4x3 board, implemented as a small program that uses a 2D boolean board, copies of the board for trial moves, and negamax depth-first recursion to back up win/loss values; the code is roughly 80–100 lines of Java (one Haskell port finishes in about an hour) and explores the ~12 first-move choices, stalls by taking the lower-right square when no forced win exists, and runs correctly with repos posted.
AI-written summary · about
chapters (7)
- 0:00Building Chomp AI
- 2:40Chomp rules and strategy
- 8:15Pseudo code AI walkthrough
- 17:30Live Haskell coding
- 1:27:55Debugging the AI
- 1:30:05Game AI strategies
- 1:36:09Q&A
related talks
57mBuilding a Chess-Playing AI with JS in 40 Minutes
08mDesigning Boardgames and How Tech (and React) Can Help
49mThat Works?! Quines and Other Delightfully Useless Programs
25mBuilding Board Games with Ruby
01h 02mThe Art of Code
01h 04mThe Art of Code
31mBattleship game
05mIt's all fun and games til your analysis code is finished