RL4F provides the first standardized offline RL benchmark for multi-actuator long-horizon plasma control using DIII-D data, with model-based methods showing best average performance across rotation, density, temperature, and pressure tasks.
R., Görler, T., and contributors, J
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Offline Reinforcement Learning for Plasma Control in Nuclear Fusion: Codebase and Benchmark
RL4F provides the first standardized offline RL benchmark for multi-actuator long-horizon plasma control using DIII-D data, with model-based methods showing best average performance across rotation, density, temperature, and pressure tasks.