English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Gradient-based meta-learning and hyperparameter optimization have seensignificant progress recently, enabling practical end-to-end training of neuralnetworks together with many hyperparameters. Nevertheless, existing approachesare relatively expensive as they need to compute second-order derivatives andstore a longer computational graph. This cost prevents scaling them to largernetwork architectures. We present EvoGrad, a new approach to meta-learning thatdraws upon evolutionary techniques to more efficiently compute hypergradients.EvoGrad estimates hypergradient with respect to hyperparameters withoutcalculating second-order gradients, or storing a longer computational graph,leading to significant improvements in efficiency. We evaluate EvoGrad on threesubstantial recent meta-learning applications, namely cross-domain few-shotlearning with feature-wise transformations, noisy label learning withMeta-Weight-Net and low-resource cross-lingual learning with metarepresentation transformation. The results show that EvoGrad significantlyimproves efficiency and enables scaling meta-learning to bigger architecturessuch as from ResNet10 to ResNet34.

EvoGrad: Efficient Gradient-Based Meta-Learning and Hyperparameter Optimization