Machine Learning for neutron source distributions
For researchers in neutron source modeling, this work introduces machine learning as a new tool for source distribution estimation, though it is an incremental application of existing generative models to a specific domain.
This paper proposes using probabilistic generative models (VAE, normalizing flow, GAN, denoising diffusion) to estimate neutron source distributions from Monte Carlo particle lists, enabling efficient and memory-less sampling after training. The models are compared to existing methods, showing that generative models can effectively model these distributions.
In light of the recent advancements in machine learning, we propose a novel approach to neutron source distribution estimation through the utilisation of probabilistic generative models. The estimation is based on a Monte Carlo particle list, which is only required during the training stage of the machine learning model. Once the source distribution has been learned, the model is independent of the original particle list, allowing for further sampling in an efficient, rapid, and memory-costless manner. The performance of various generative models is evaluated, including a variational autoencoder, a normalizing flow, a generative adversarial network, and a denoising diffusion model. These approaches are then compared to existing source distribution estimations, and the advantages and disadvantages of each approach are discussed. The results demonstrate that source distributions can be modeled through the use of probabilistic generative models, which paves the way for further advancements in this field.