We lifted 2 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| facebookresearch/rebel | canonical | 2 of 2 |
| Function | Status | Where it lives |
|---|---|---|
| clip_grad_norm_ | Ran | facebookresearch/rebel/cfvpy/selfplay.py code served (permissive licence) · get_code("6300179d0f874eb5") |
| get_last_action_index | Ran | facebookresearch/rebel/cfvpy/selfplay.py code served (permissive licence) · get_code("a1fb618ab3092a50") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
The combination of deep reinforcement learning and search at both training and test time is a powerful paradigm that has led to a number of successes in single-agent settings and perfect-information games, best exemplified by AlphaZero. However, prior algorithms of this form cannot cope with imperfect-information games. This paper presents ReBeL, a general framework for self-play reinforcement learning and search that provably converges to a Nash equilibrium in any two-player zero-sum game. In the simpler setting of perfect-information games, ReBeL reduces to an algorithm similar to AlphaZero. Results in two different imperfect-information games show ReBeL converges to an approximate Nash equilibrium. We also show ReBeL achieves superhuman performance in heads-up no-limit Texas hold'em poker, while using far less domain knowledge than any prior poker AI.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2007.13544")
get_code_for_paper("2007.13544")
have("2007.13544")
Connect an agent — have() is free.