Monte Carlo reinforcement learning on the canonical ordering of World 1-1 segments converges fastest with highest efficiency and no catastrophic failures, unlike any random permutation.
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
-
Reinforcement Learning in Super Mario Bros: Curriculum, Pedagogy, and Optimal Level Design in World 1-1
Monte Carlo reinforcement learning on the canonical ordering of World 1-1 segments converges fastest with highest efficiency and no catastrophic failures, unlike any random permutation.