/hw7

Primary LanguageGo

やったこと

min max (pull-req: modify-min-max)

depth=3 にして，greedyとrandomには勝てるようになった

alpha-beta (pull-req: modify-min-max)

これでdepth=3 以上も試せると思ったら，depth=4でも時間切れになってしまう．
実装を見直したけれど，やはりdepth=4でも時間切れになる．．．

alpha-beta + 盤面の評価値の変更 (pull-req: add-new-score-list)

初めのうちは角を取ろうと頑張るようにした．
うまくいくかな．

困っていること

  • alpha-beta法の実装間違っているのかな． => プリントしながらdepth=1, 2を見てみたけれど，合っていそう．
  • 賢そうな挙動はあまりしていない． => 戦略を増やす．
  • jsonの中身の，historyとかの取り出し方が分からない．

Homework 7

  1. Fork a copy for your changes
  2. Clone it to your local workstation
  3. Create a new project to host your AI in https://console.cloud.google.com
  4. Run gcloud init and select that new project.
  5. gcloud app deploy python/ or gcloud app deploy go/ to deploy your app.
  6. Add the appspot address to the "Reversi Players" sheet
  7. Modify the way a move is picked
  8. re-deploy the app
  9. repeat steps 7 and 8 until you have a very clever AI :)
  10. eventually push your awesome clever AI to github.
    • If you want to keep it secret until Thursday night, that's fine.
  11. email step17 with your github repository link.

Using reflector.go

You can use this "reflector" program to make a locally running dev_appserver instance act like a human player (i.e. you don't have to deploy the whole app to have it run a game).

To run it: