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

The unstructured sparsity after pruning poses a challenge to the efficientimplementation of deep learning models in existing regular architectures likesystolic arrays. On the other hand, coarse-grained structured pruning issuitable for implementation in regular architectures but tends to have higheraccuracy loss than unstructured pruning when the pruned models are of the samesize. In this work, we propose a model compression method based on a novelweight permutation scheme to fully exploit the fine-grained weight sparsity inthe hardware design. Through permutation, the optimal arrangement of the weightmatrix is obtained, and the sparse weight matrix is further compressed to asmall and dense format to make full use of the hardware resources. Two pruninggranularities are explored. In addition to the unstructured weight pruning, wealso propose a more fine-grained subword-level pruning to further improve thecompression performance. Compared to the state-of-the-art works, the matrixcompression rate is significantly improved from 5.88x to 14.13x. As a result,the throughput and energy efficiency are improved by 2.75 and 1.86 times,respectively.

Tight Compression: Compressing CNN Through Fine-Grained Pruning and Weight Permutation for Efficient Implementation