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

We propose precision gating (PG), an end-to-end trainable dynamicdual-precision quantization technique for deep neural networks. PG computesmost features in a low precision and only a small proportion of importantfeatures in a higher precision to preserve accuracy. The proposed approach isapplicable to a variety of DNN architectures and significantly reduces thecomputational cost of DNN execution with almost no accuracy loss. Ourexperiments indicate that PG achieves excellent results on CNNs, includingstatically compressed mobile-friendly networks such as ShuffleNet. Compared tothe state-of-the-art prediction-based quantization schemes, PG achieves thesame or higher accuracy with 2.4$\times$ less compute on ImageNet. PGfurthermore applies to RNNs. Compared to 8-bit uniform quantization, PG obtainsa 1.2% improvement in perplexity per word with 2.7$\times$ computational costreduction on LSTM on the Penn Tree Bank dataset. Code is available at:https://github.com/cornell-zhang/dnn-gating

Precision Gating: Improving Neural Network Efficiency with Dynamic Dual-Precision Activations