Interpretable Meta-Learning of Physical Systems

arXiv:2312.00477v2 Announce Type: replace
Abstract: Machine learning methods can be a valuable aid in the scientific process, but they need to face challenging settings where data come from inhomogeneous experimental conditions. Recent meta-learning methods have made significant progress in multi-task learning, but they rely on black-box neural networks, resulting in high computational costs and limited interpretability. Leveraging the structure of the learning problem, we argue that multi-environment generalization can be achieved using a simpler learning model, with an affine structure …

abstract arxiv box computational costs cs.lg data experimental face machine machine learning meta meta-learning multi-task learning networks neural networks process progress scientific stat.ml systems type