Multi-Timescale Dynamics Model Bayesian Optimization for Plasma Stabilization in Tokamaks

Rohit Sonker, Alexandre Capone, Andrew Rothstein, Hiro Josep Farre Kaga, Egemen Kolemen, Jeff Schneider
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:56465-56479, 2025.

Abstract

Machine learning algorithms often struggle to control complex real-world systems. In the case of nuclear fusion, these challenges are exacerbated, as the dynamics are notoriously complex, data is poor, hardware is subject to failures, and experiments often affect dynamics beyond the experiment’s duration. Existing tools like reinforcement learning, supervised learning, and Bayesian optimization address some of these challenges but fail to provide a comprehensive solution. To overcome these limitations, we present a multi-scale Bayesian optimization approach that integrates a high-frequency data-driven dynamics model with a low-frequency Gaussian process. By updating the Gaussian process between experiments, the method rapidly adapts to new data, refining the predictions of the less reliable dynamical model. We validate our approach by controlling tearing instabilities in the DIII-D nuclear fusion plant. Offline testing on historical data shows that our method significantly outperforms several baselines. Results on live experiments on the DIII-D tokamak, conducted under high-performance plasma scenarios prone to instabilities, shows a 50% success rate — marking a 117% improvement over historical outcomes.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-sonker25a, title = {Multi-Timescale Dynamics Model {B}ayesian Optimization for Plasma Stabilization in Tokamaks}, author = {Sonker, Rohit and Capone, Alexandre and Rothstein, Andrew and Kaga, Hiro Josep Farre and Kolemen, Egemen and Schneider, Jeff}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {56465--56479}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/sonker25a/sonker25a.pdf}, url = {https://proceedings.mlr.press/v267/sonker25a.html}, abstract = {Machine learning algorithms often struggle to control complex real-world systems. In the case of nuclear fusion, these challenges are exacerbated, as the dynamics are notoriously complex, data is poor, hardware is subject to failures, and experiments often affect dynamics beyond the experiment’s duration. Existing tools like reinforcement learning, supervised learning, and Bayesian optimization address some of these challenges but fail to provide a comprehensive solution. To overcome these limitations, we present a multi-scale Bayesian optimization approach that integrates a high-frequency data-driven dynamics model with a low-frequency Gaussian process. By updating the Gaussian process between experiments, the method rapidly adapts to new data, refining the predictions of the less reliable dynamical model. We validate our approach by controlling tearing instabilities in the DIII-D nuclear fusion plant. Offline testing on historical data shows that our method significantly outperforms several baselines. Results on live experiments on the DIII-D tokamak, conducted under high-performance plasma scenarios prone to instabilities, shows a 50% success rate — marking a 117% improvement over historical outcomes.} }
Endnote
%0 Conference Paper %T Multi-Timescale Dynamics Model Bayesian Optimization for Plasma Stabilization in Tokamaks %A Rohit Sonker %A Alexandre Capone %A Andrew Rothstein %A Hiro Josep Farre Kaga %A Egemen Kolemen %A Jeff Schneider %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-sonker25a %I PMLR %P 56465--56479 %U https://proceedings.mlr.press/v267/sonker25a.html %V 267 %X Machine learning algorithms often struggle to control complex real-world systems. In the case of nuclear fusion, these challenges are exacerbated, as the dynamics are notoriously complex, data is poor, hardware is subject to failures, and experiments often affect dynamics beyond the experiment’s duration. Existing tools like reinforcement learning, supervised learning, and Bayesian optimization address some of these challenges but fail to provide a comprehensive solution. To overcome these limitations, we present a multi-scale Bayesian optimization approach that integrates a high-frequency data-driven dynamics model with a low-frequency Gaussian process. By updating the Gaussian process between experiments, the method rapidly adapts to new data, refining the predictions of the less reliable dynamical model. We validate our approach by controlling tearing instabilities in the DIII-D nuclear fusion plant. Offline testing on historical data shows that our method significantly outperforms several baselines. Results on live experiments on the DIII-D tokamak, conducted under high-performance plasma scenarios prone to instabilities, shows a 50% success rate — marking a 117% improvement over historical outcomes.
APA
Sonker, R., Capone, A., Rothstein, A., Kaga, H.J.F., Kolemen, E. & Schneider, J.. (2025). Multi-Timescale Dynamics Model Bayesian Optimization for Plasma Stabilization in Tokamaks. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:56465-56479 Available from https://proceedings.mlr.press/v267/sonker25a.html.

Related Material