Uncertainty based Online Ensemble on Non-Stationary Data for Fusion Science

Conference · Sat Dec 06 00:00:00 EST 2025

OSTI ID:3012386

Rajput, Kishansingh ^[1]; Sammuli, Brian; Schram, Malachi ^[1]; Lin, Sen

---

1. Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States)

Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior due to drifts in the data. The drifts can arise from both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with non-stationary data streams.Online learning can be used to continuously adapt the models with new data as it is acquired. However, traditional online learning can suffer from short-term performance degradation, as ground truth are not available before making the prediction. To address this challenge, we propose uncertainty aware ensemble approach for online learning. We use Deep Gaussian Process Approximation (DGPA) technique for calibrated uncertainty estimation and use the uncertainty values to guide a meta-algorithm that produces predictions based on ensemble of learners. Moreover, DGPA also provides uncertainty estimation along with the predictions for decision makers. This paper demonstrates that the proposed method outperforms traditional online learning approach, and a naive ensemble without uncertainty guidance by about 7% and 6%, respectively, on B-coil deflection prediction at DIII-D Fusion Facility.

• Research Organization: Thomas Jefferson National Accelerator Facility (TJNAF)
• Sponsoring Organization: USDOE Office of Science (SC), Nuclear Physics (NP)
• DOE Contract Number: AC05-06OR23177
• OSTI ID: 3012386
• Report Number(s): JLAB-CST-25-4636; DOE/OR/23177-8093
• Resource Type: Conference paper
• Conference Information: Machine Learning and the Physical Sciences, Workshop at the 39th conference on Neural Information Processing Systems (NeurIPS), Saturday, December 6, 2025
• Country of Publication: United States
• Language: English