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

Well-trained deep neural networks (DNNs) treat all test samples equallyduring prediction. Adaptive DNN inference with early exiting leverages theobservation that some test examples can be easier to predict than others. Thispaper presents EENet, a novel early-exiting scheduling framework for multi-exitDNN models. Instead of having every sample go through all DNN layers duringprediction, EENet learns an early exit scheduler, which can intelligentlyterminate the inference earlier for certain predictions, which the model hashigh confidence of early exit. As opposed to previous early-exiting solutionswith heuristics-based methods, our EENet framework optimizes an early-exitingpolicy to maximize model accuracy while satisfying the given per-sample averageinference budget. Extensive experiments are conducted on four computer visiondatasets (CIFAR-10, CIFAR-100, ImageNet, Cityscapes) and two NLP datasets(SST-2, AgNews). The results demonstrate that the adaptive inference by EENetcan outperform the representative existing early exit techniques. We alsoperform a detailed visualization analysis of the comparison results tointerpret the benefits of EENet.

Adaptive Deep Neural Network Inference Optimization with EENet