A Globally Convergent Algorithm for Neural Network Parameter Optimization Based on Difference-of-Convex Functions
Open access
Date
2024-01Type
- Journal Article
ETH Bibliography
yes
Altmetrics
Abstract
We propose an algorithm for optimizing the parameters of single hidden layer neural networks. Specifically, we derive a blockwise difference-of-convex (DC) functions representation of the objective function. Based on the latter, we propose a block coordinate descent (BCD) approach that we combine with a tailored difference-of-convex functions algorithm (DCA). We prove global convergence of the proposed algorithm. Furthermore, we mathematically analyze the convergence rate of parameters and the convergence rate in value (i.e., the training loss). We give conditions under which our algorithm converges linearly or even faster depending on the local shape of the loss function. We confirm our theoretical derivations numerically and compare our algorithm against state-of-the-art gradient-based solvers in terms of both training loss and test loss. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000652380Publication status
publishedExternal links
Journal / series
Transactions on Machine Learning ResearchVolume
Publisher
OpenReviewOrganisational unit
09623 - Feuerriegel, Stefan (ehemalig) / Feuerriegel, Stefan (former)
More
Show all metadata
ETH Bibliography
yes
Altmetrics