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

While selecting the hyper-parameters of Neural Networks (NNs) has been so fartreated as an art, the emergence of more complex, deeper architectures posesincreasingly more challenges to designers and Machine Learning (ML)practitioners, especially when power and memory constraints need to beconsidered. In this work, we propose HyperPower, a framework that enablesefficient Bayesian optimization and random search in the context of power- andmemory-constrained hyper-parameter optimization for NNs running on a givenhardware platform. HyperPower is the first work (i) to show that powerconsumption can be used as a low-cost, a priori known constraint, and (ii) topropose predictive models for the power and memory of NNs executing on GPUs.Thanks to HyperPower, the number of function evaluations and the best testerror achieved by a constraint-unaware method are reached up to 112.99x and30.12x faster, respectively, while never considering invalid configurations.HyperPower significantly speeds up the hyper-parameter optimization, achievingup to 57.20x more function evaluations compared to constraint-unaware methodsfor a given time interval, effectively yielding significant accuracyimprovements by up to 67.6%.

HyperPower: Power- and Memory-Constrained Hyper-Parameter Optimization for Neural Networks