The paper presents a PPO policy trained with rewards from a learned cost model, claiming 99.4% success on the 2x2x2 Rubik's Cube without search or solved-state sampling, but with weak evidential support.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Solving Rubik's Cube Without Tricky Sampling
The paper presents a PPO policy trained with rewards from a learned cost model, claiming 99.4% success on the 2x2x2 Rubik's Cube without search or solved-state sampling, but with weak evidential support.