The Leela Chess Zero policy network encodes information about destination squares of moves up to seven plies ahead, with attention heads that copy future-square information backward in time in a pattern-dependent way.
Nonetheless, the contribution of L12H12 is relatively limited
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Understanding the learned look-ahead behavior of chess neural networks
The Leela Chess Zero policy network encodes information about destination squares of moves up to seven plies ahead, with attention heads that copy future-square information backward in time in a pattern-dependent way.