PolyThrottle: Energy-efficient Neural Network Inference on Edge Devices: Opportunities

  • 📰 hackernoon
  • ⏱ Reading Time:
  • 24 sec. here
  • 2 min. at publisher
  • 📊 Quality Score:
  • News: 13%
  • Publisher: 51%

Portugal Manchetes Notícia

Portugal Últimas Notícias,Portugal Manchetes

This paper investigates how the configuration of on-device hardware affects energy consumption for neural network inference with regular fine-tuning.

This paper is available on arxiv under CC BY-NC-ND 4.0 DEED license. Authors: Minghao Yan, University of Wisconsin-Madison; Hongyi Wang, Carnegie Mellon University; Shivaram Venkataraman, myan@cs.wisc.edu. Table of Links Abstract & Introduction Motivation Opportunities Architecture Overview Proble Formulation: Two-Phase Tuning Modeling Workload Interference Experiments Conclusion & References A. Hardware Details B. Experimental Results C. Arithmetic Intensity D.

This indicates that the Results default memory frequency is higher than optimal for modern Deep Learning workloads. For heavy workloads such as Bert, memory tuning can account for the majority of the energy consumption reduction. This can be partially attributed to the memory-bound nature of Transformer-based models . Our result demonstrates that systems that aim to optimize energy use in neural network inference need to take memory frequency into account.

 

Obrigado pelo seu comentário. Seu comentário será publicado após ser revisado.
Resumimos esta notícia para que você possa lê-la rapidamente. Se você se interessou pela notícia, pode ler o texto completo aqui. Consulte Mais informação:

 /  🏆 532. in PT

Portugal Últimas Notícias, Portugal Manchetes

Similar News:Você também pode ler notícias semelhantes a esta que coletamos de outras fontes de notícias.

PolyThrottle: Energy-efficient Neural Network Inference on Edge Devices: Experimental ResultsThis paper investigates how the configuration of on-device hardware affects energy consumption for neural network inference with regular fine-tuning.
Fonte: hackernoon - 🏆 532. / 51 Consulte Mais informação »

Using Machine Learning & Neural Networks to Maximize Solar Energy GloballyOther ways to generate solar energy that just trying to make solar cells efficient potential of luminescent solar concentration
Fonte: cleantechnica - 🏆 565. / 51 Consulte Mais informação »

CPS Energy progressing towards cleaner energy goals; set to shut down older plantsSAN ANTONIO- Cleaner energy: that's the goal cps energy is working toward -- by closing down coal and older natural gas plants. CPS energy announced they will b
Fonte: News4SA - 🏆 251. / 63 Consulte Mais informação »

Yellen: Pushing Green Energy to Lower Energy Costs ‘Over Time’ Is Key Part of Fighting InflationSource of breaking news and analysis, insightful commentary and original reporting, curated and written specifically for the new generation of independent and conservative thinkers.
Fonte: BreitbartNews - 🏆 610. / 51 Consulte Mais informação »

$22 Million to Improve Siting & Permitting for Large-Scale Renewable Energy & Energy StorageInvests $22 Million to Improve Siting and Permitting for Large-Scale Renewable Energy and Energy Storage in NC
Fonte: cleantechnica - 🏆 565. / 51 Consulte Mais informação »