-
https://github.com/mokemokechicken/reversi-alpha-zero/blob/f1cfa6c7177ec5f76a89e20fd97eb4c5d678611d/src/reversi_zero/agent/player.py#L165-L168
I see update N and W with virtual loss when select the…
-
Thanks for your nice wok!
I set BENCHMARK_DIR in both `main.py` as `/home/sun/Desktop/ABC-RL/arithmetic` ;
I set HOME_DIR in `mcts_agent_training.sh` as `/home/sun/Desktop/ABC-RL`
While runi…
-
Hi
I have ran the search, rollout and select_next_move methods but im still not sure how to code the adapt method:
1. In line 52 and 53, what does possible move mean? Is it the move a stop could ha…
-
Hi NeverOnTimeSdnBhd,
There are 2 questions I am unsure of and require your explanation,
1. Is it okay to use level = 3, iterations = 100 in each sample test case?
2. What does the 3D array of poli…
-
1) 3d array initialization
My code is almost the same with Issue #11 (Sir said it was correct), but we got a ArrayIndexOutOfBoundsException error.
I have checked and tried for many times and the er…
-
Smooth UCB:
[Self-Play Monte-Carlo Tree Search in Computer Poker]
https://pdfs.semanticscholar.org/7b68/7599b4425aa959036071030e1212a3b359c7.pdf
-
-
A result of #50
#51 is a dependency
-
Hi! I just went over mcts.py. Here are few remarks:
- Great work :) this looks nice!
- The `MCTSNode` attributes `prior` and `policy` seems redundant: `node.prior = node.parent._policy['child_inde…
-
MCTS player crashes in 10x10 board--fixed