@@ -49,27 +49,27 @@ board for the playouts from a given node ::
49
49
50
50
These AI players can also take additional arguments:
51
51
52
- time
52
+ time (default: 30)
53
53
The amount of thinking time allowed for the AI to make its decision,
54
- in seconds (default: 30). Ex: ``$ board-play.py t3 jrb.mcts.uct -e
55
- time=5 ``
54
+ in seconds. Ex: ``$ board-play.py t3 jrb.mcts.uct -e time=5 ``
56
55
57
- max_actions
56
+ max_actions (default: 1000)
58
57
The maximum number of actions, or plays, to allow in one of the
59
- simulated playouts before giving up (default: 1000) . Ex: ``$
60
- board-play.py t3 jrb.mcts.uct -e max_actions=500 ``
58
+ simulated playouts before giving up. Ex: ``$ board-play.py t3
59
+ jrb.mcts.uct -e max_actions=500 ``
61
60
62
- C
61
+ C (default: 1.4)
63
62
The exploration vs. exploitation coefficient at the heart of the UCT
64
- algorithm (default: 1.4). Larger values prioritize exploring
65
- inadequately covered actions from a node, smaller values prioritize
66
- exploiting known higher valued actions. Experimentation with this
67
- variable to find reasonable values for a given game is recommended.
68
- Ex: `` $ board-play.py t3 jrb.mcts.uct -e C=3.5 ``
63
+ algorithm. Larger values prioritize exploring inadequately covered
64
+ actions from a node, smaller values prioritize exploiting known
65
+ higher valued actions. Experimentation with this variable to find
66
+ reasonable values for a given game is recommended. Ex: `` $
67
+ board-play.py t3 jrb.mcts.uct -e C=3.5 ``
69
68
70
69
The ``-e `` flag may be used multiple times to set additional
71
70
variables.
72
71
72
+
73
73
Games
74
74
-----
75
75
0 commit comments