Fast and Efficient Local Search for Genetic Programming Based Loss Function Learning

arXiv:2403.00865v1 Announce Type: cross
Abstract: In this paper, we develop upon the topic of loss function learning, an emergent meta-learning paradigm that aims to learn loss functions that significantly improve the performance of the models trained under them. Specifically, we propose a new meta-learning framework for task and model-agnostic loss function learning via a hybrid search approach. The framework first uses genetic programming to find a set of symbolic loss functions. Second, the set of learned loss functions is subsequently …

abstract arxiv cs.ai cs.cv cs.lg cs.ne framework function functions genetic programming learn loss meta meta-learning paper paradigm performance programming search them type