Open mitchellfaulk opened 1 year ago
Such a model would include the quasi-thrify/quasi-greedy as the depth=1 case, but I want to be able to go deeper. The depth=2 case should be related to the thrifty/greedy algorithms from the setgametools.py script.
Design a tree-based model whose future states are evaluated only on the current information (and probabilities concerning future states)