GPU Power Capping for Energy-Performance Trade-Offs in Training of Deep Convolutional Neural Networks for Image Recognition.
Adam KrzywaniakPawel CzarnulJerzy ProficzPublished in: ICCS (1) (2022)
Keyphrases
- image recognition
- convolutional neural networks
- trade off
- image classification
- pattern recognition
- energy management
- power consumption
- real time
- face recognition
- training set
- deep architectures
- energy consumption
- electrical energy
- convolutional network
- tchebichef moments
- graphics hardware
- energy efficiency
- general purpose
- supervised learning
- feature vectors
- multithreading
- power generation
- total energy
- energy saving
- neural network