Evolving Energy Efficient Convolutional Neural Networks
- ORNL
As deep neural networks have been deployed in more and more applications over the past half decade and are finding their way into an ever increasing number of operational systems, their energy consumption becomes a concern whether running in the datacenter or on edge devices. Hyperparameter optimization and automated network design for deep learning is a quickly growing field, but much of the focus has remained only on optimizing for the performance of the machine learning task. In this work, we demonstrate that the best performing networks created through this automated network design process have radically different computational characteristics (e.g. energy usage, model size, inference time), presenting the opportunity to utilize this optimization process to make deep learning networks more energy efficient and deployable to smaller devices. Optimizing for these computational characteristics is critical as the number of applications of deep learning continues to expand.
- Research Organization:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- DOE Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1606807
- Resource Relation:
- Conference: 2nd Workshop on Energy-Efficient Machine Learning and Big Data Analytics (in conjuction with IEEE Big Data) - Los Angeles, California, United States of America - 11/9/2019 5:00:00 AM-11/12/2019 5:00:00 AM
- Country of Publication:
- United States
- Language:
- English
Similar Records
IL-Net: Using Expert Knowledge to Guide the Design of Furcated Neural Networks
Convolutional Neural Networks for the CHIPS Neutrino Detector R&D Project.