CoRe Optimizer: An All-in-One Solution for Machine Learning

arXiv:2307.15663v2 Announce Type: replace
Abstract: The optimization algorithm and its hyperparameters can significantly affect the training speed and resulting model accuracy in machine learning applications. The wish list for an ideal optimizer includes fast and smooth convergence to low error, low computational demand, and general applicability. Our recently introduced continual resilient (CoRe) optimizer has shown superior performance compared to other state-of-the-art first-order gradient-based optimizers for training lifelong machine learning potentials. In this work we provide an extensive performance comparison of …

abstract accuracy algorithm applications arxiv computational continual convergence core cs.lg demand error general list low machine machine learning machine learning applications math.oc model accuracy optimization physics.comp-ph resilient solution speed training type