Can Kans (re)discover predictive models for Direct-Drive Laser Fusion?
This work addresses predictive modeling for laser fusion energy, an incremental improvement in a domain-specific application with data constraints.
The paper tackled the challenge of predictive modeling in laser fusion, where high complexity and limited data exist, by using Kolmogorov-Arnold Networks (KANs) as an alternative to physics-informed learning, achieving high prediction accuracy and interpretability compared to other models.
The domain of laser fusion presents a unique and challenging predictive modeling application landscape for machine learning methods due to high problem complexity and limited training data. Data-driven approaches utilizing prescribed functional forms, inductive biases and physics-informed learning (PIL) schemes have been successful in the past for achieving desired generalization ability and model interpretation that aligns with physics expectations. In complex multi-physics application domains, however, it is not always obvious how architectural biases or discriminative penalties can be formulated. In this work, focusing on nuclear fusion energy using high powered lasers, we present the use of Kolmogorov-Arnold Networks (KANs) as an alternative to PIL for developing a new type of data-driven predictive model which is able to achieve high prediction accuracy and physics interpretability. A KAN based model, a MLP with PIL, and a baseline MLP model are compared in generalization ability and interpretation with a domain expert-derived symbolic regression model. Through empirical studies in this high physics complexity domain, we show that KANs can potentially provide benefits when developing predictive models for data-starved physics applications.