0.25
0.5
0.75
1.25
1.5
1.75
2
Ensemble Monte-Carlo Planning: An Empirical Study
Published on Jul 21, 20114288 Views
Monte-Carlo planning algorithms, such as UCT, select actions at each decision epoch by intelligently expanding a single search tree given the available time and then selecting the best root action. Re
Related categories
Chapter list
Ensemble Monte-Carlo Planning: An Empirical Study00:00
Talk Outline - 100:12
Klondike Solitaire00:31
Parallel UCT in Go02:49
Prior Observations: Multi-Core04:30
Prior Observations: Single-Core05:00
Objective06:17
Talk Outline - 207:02
UCT Algorithm07:12
Monte-Carlo Tree Search07:59
UCT Example - 109:16
UCT Example - 210:02
UCT Example - 310:13
UCT Example - 410:20
UCT Example - 510:41
UCT Example - 610:45
UCT Example - 710:46
UCT Example - 810:46
UCT Example - 910:53
Talk Outline - 311:20
Ensemble UCT11:23
Why might ensembles work? - 112:23
Why might ensembles work? - 213:07
Why might ensembles work? - 313:48
Talk Outline - 415:03
Backgammon15:07
Biniax15:59
Connect 416:49
Havannah17:21
Yahtzee17:54
Experiment Setup18:20
Results - 119:28
Results - 219:53
Results - 320:11
Results - 420:12
Results - 520:44
Results - 621:10
Results - 721:12
Results - 821:37
Results: Single Core21:53
Results - 922:20
Results - 1022:46
Small Trees22:57
Results - 1123:08
Summary23:39
Future Work24:06
Thanks24:38