> Abstract:Machine learning models are increasingly used in feedback control loops for nuclear fusion, where inference speed and predictable timing are critical. We summarize lessons from models deployed for control on the DIII-D tokamak and develop a benchmark to compare inference backends across ten neural networks and model components from fusion control and diagnostic pipelines. For models greater than five million parameters, the CPU backends take tens to thousands of milliseconds, while GPU inference is substantially faster, suggesting an upper limit on CPU-oriented development for control. These results show why the deployment backend must be selected together with the model and its control-cycle budget.
| | | | --- | --- | | Subjects: | Plasma Physics (physics.plasm-ph); Performance (cs.PF); Systems and Control (eess.SY) | | Cite as: | arXiv:2610.00845 \[physics.plasm-ph\] | | | (or arXiv:2610.00845v1 \[physics.plasm-ph\] for this version) | | | https://doi.org/10.48550/arXiv.2610.00845 Focus to learn more arXiv-issued DOI via DataCite (pending registration) |
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.