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

Unstructured pruning is well suited to reduce the memory footprint ofconvolutional neural networks (CNNs), both at training and inference time. CNNscontain parameters arranged in $K \times K$ filters. Standard unstructuredpruning (SP) reduces the memory footprint of CNNs by setting filter elements tozero, thereby specifying a fixed subspace that constrains the filter.Especially if pruning is applied before or during training, this induces astrong bias. To overcome this, we introduce interspace pruning (IP), a generaltool to improve existing pruning methods. It uses filters represented in adynamic interspace by linear combinations of an underlying adaptive filterbasis (FB). For IP, FB coefficients are set to zero while un-prunedcoefficients and FBs are trained jointly. In this work, we provide mathematicalevidence for IP's superior performance and demonstrate that IP outperforms SPon all tested state-of-the-art unstructured pruning methods. Especially inchallenging situations, like pruning for ImageNet or pruning to high sparsity,IP greatly exceeds SP with equal runtime and parameter costs. Finally, we showthat advances of IP are due to improved trainability and superiorgeneralization ability.

Interspace Pruning: Using Adaptive Filter Representations to Improve Training of Sparse CNNs