A transition graph model with utility and evidence counts learns behaviors from state history and feedback, showing performance comparable to neural networks on Atari Breakout.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Interpretable experiential learning based on state history and global feedback
A transition graph model with utility and evidence counts learns behaviors from state history and feedback, showing performance comparable to neural networks on Atari Breakout.