Neural Loss Function Evolution for Large-Scale Image Classifier Convolutional Neural Networks

arXiv:2403.08793v1 Announce Type: cross
Abstract: For classification, neural networks typically learn by minimizing cross-entropy, but are evaluated and compared using accuracy. This disparity suggests neural loss function search (NLFS), the search for a drop-in replacement loss function of cross-entropy for neural networks. We apply NLFS to image classifier convolutional neural networks. We propose a new search space for NLFS that encourages more diverse loss functions to be explored, and a surrogate function that accurately transfers to large-scale convolutional neural networks. …

abstract accuracy apply arxiv classification classifier convolutional neural networks cross-entropy cs.cv cs.lg entropy evolution function image learn loss networks neural networks replacement scale search type