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 state-of-art DNN structures involve intensive computation and high memorystorage. To mitigate the challenges, the memristor crossbar array has emergedas an intrinsically suitable matrix computation and low-power accelerationframework for DNN applications. However, the high accuracy solution for extrememodel compression on memristor crossbar array architecture is still waiting forunraveling. In this paper, we propose a memristor-based DNN framework whichcombines both structured weight pruning and quantization by incorporatingalternating direction method of multipliers (ADMM) algorithm for better pruningand quantization performance. We also discover the non-optimality of the ADMMsolution in weight pruning and the unused data path in a structured prunedmodel. Motivated by these discoveries, we design a software-hardwareco-optimization framework which contains the first proposed NetworkPurification and Unused Path Removal algorithms targeting on post-processing astructured pruned model after ADMM steps. By taking memristor hardwareconstraints into our whole framework, we achieve extreme high compression ratioon the state-of-art neural network structures with minimum accuracy loss. Forquantizing structured pruned model, our framework achieves nearly no accuracyloss after quantizing weights to 8-bit memristor weight representation. Weshare our models at anonymous link https://bit.ly/2VnMUy0.

Tiny but Accurate: A Pruned, Quantized and Optimized Memristor Crossbar Framework for Ultra Efficient DNN Implementation