Transfer learning for analysis of collective and non-collective Thomson scattering spectra

Thomson scattering (TS) diagnostics provide reliable, minimally perturbative measurements of fundamental plasma parameters, such as electron density (⁠n_e) and electron temperature (⁠T_e⁠). Deep neural networks can provide accurate estimates of ⁠n_e and T_e when conventional fitting algorithms may fail, such as when TS spectra are dominated by noise, or when fast analysis is required for real-time operation. Although deep neural networks typically require large training sets, transfer learning can improve model performance on a target task with limited data by leveraging pre-trained models from related source tasks, where select hidden layers are further trained using target data. We present five architecturally diverse deep neural networks, pre-trained on synthetic TS data and adapted for experimentally measured TS data, to evaluate the efficacy of transfer learning in estimating n_e and T_e in both the collective and non-collective scattering regimes. We evaluate errors in n_e and T_e estimates as a function of training set size for models trained with and without transfer learning, and we observe decreases in model error from transfer learning when the training set contains ≲ 200 experimentally measured spectra.. Authors: Van Hoomissen, T. [University of California—Los Angeles, CA (United States)] (ORCID:0009000772056926); Alhuthali, J. [University of California—Los Angeles, CA (United States)]; Ortiz, A. M. [University of California—Los Angeles, CA (United States)]; Mariscal, D. A. [Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)] (ORCID:0000000195024205); Dorst, R. S. [University of California—Los Angeles, CA (United States)]. DOE Contract: AC52-07NA27344; NA0003856; NA0004033; NA0004147; SC0020431; SC0024549. Subjects: Artificial neural networks; Convolutional neural network; Machine learning; Optimization algorithms; Plasma waves; Shock waves; Thomson scattering