Reinforcement learning finds energy-efficient paths across convective cells by learning to cross flow barriers and ride attracting structures, with performance improving as Rayleigh number rises.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.flu-dyn 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning to traverse convective flows at moderate to high Rayleigh numbers
Reinforcement learning finds energy-efficient paths across convective cells by learning to cross flow barriers and ride attracting structures, with performance improving as Rayleigh number rises.