mlsanityData to 14 Oct 2026

Plasma Shape Control via Zero-shot Generative Reinforcement Learning

arXiv 2510.17531 ↗First posted 20 Oct 2025physics.plasm-ph cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL for energy, networks and operations

Abstract (arXiv)

Traditional PID controllers have limited adaptability for plasma shape control, and task-specific reinforcement learning (RL) methods suffer from limited generalization and the need for repetitive retraining. To overcome these challenges, this paper proposes a novel framework for developing a versatile, zero-shot control policy from a large-scale offline dataset of historical PID-controlled discharges. Our approach synergistically combines Generative Adversarial Imitation Learning (GAIL) with Hilbert space representation learning to achieve dual objectives: mimicking the stable operational style of the PID data and constructing a geometrically structured latent space for efficient, goal-directed control. The resulting foundation policy can be deployed for diverse trajectory tracking tasks in a zero-shot manner without any task-specific fine-tuning. Evaluations on the HL-3 tokamak simulator demonstrate that the policy excels at precisely and stably tracking reference trajectories for key shape parameters across a range of plasma scenarios. This work presents a viable pathway toward developing highly flexible and data-efficient intelligent control systems for future fusion reactors.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for energy, networks and operations, one of 15 directions in Deep reinforcement learning methods. Builds on 12 of its 42 references that are in the map. First posted on arXiv 20 Oct 2025.

Builds on

12 of 42 references.

References · 12

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2017Improved Training of Wasserstein GANscited by 2,235
  4. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
  5. 2016Generative Adversarial Imitation Learningcited by 1,597
  6. 2016Least Squares Generative Adversarial Networkscited by 1,001
  7. 2022A Generalist Agentcited by 808
  8. 2020Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPOcited by 186
  9. 2023CALM: Conditional Adversarial Latent Models for Directable Virtual Characterscited by 94
  10. 2024Foundation Policies with Hilbert Representationscited by 94
  11. 2023Lifelike Agility and Play in Quadrupedal Robots using Reinforcement Learning and Generative Pre-trained Modelscited by 37
  12. 2022Policy Adaptation from Foundation Model Feedbackcited by 10

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Dynamic Plasma Shape Control with Arbitrary Sensor Subsets ↗
  2. 2026Offline Reinforcement Learning for Plasma Control in Nuclear Fusion: Codebase and Benchmark
  3. 2026Offline Reinforcement Learning for Rotation Profile Control in Tokamaks
  4. 2025Gym-TORAX: Open-source software for integrating reinforcement learning with plasma control simulators in tokamak research
  5. 2026Provable imitation learning for control of instability in partially-observed Vlasov--Poisson equations
  6. 2024Active Disruption Avoidance and Trajectory Design for Tokamak Ramp-downs with Neural Differential Equations and Reinforcement Learning
  7. 2025Learning Plasma Dynamics and Robust Rampdown Trajectories with Predict-First Experiments at TCV
  8. 2025Enabling Integrated AI Control on DIII-D: A Control System Design with State-of-the-art Experiments