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 intrinsic error tolerance of neural network (NN) makes approximatecomputing a promising technique to improve the energy efficiency of NNinference. Conventional approximate computing focuses on balancing theefficiency-accuracy trade-off for existing pre-trained networks, which can leadto suboptimal solutions. In this paper, we propose AxTrain, a hardware-orientedtraining framework to facilitate approximate computing for NN inference.Specifically, AxTrain leverages the synergy between two orthogonalmethods---one actively searches for a network parameters distribution with higherror tolerance, and the other passively learns resilient weights bynumerically incorporating the noise distributions of the approximate hardwarein the forward pass during the training phase. Experimental results fromvarious datasets with near-threshold computing and approximation multiplicationstrategies demonstrate AxTrain's ability to obtain resilient neural networkparameters and system energy efficiency improvement.

AxTrain: Hardware-Oriented Neural Network Training for Approximate Inference