Hyperparameter Optimization in Binary Communication Networks for Neuromorphic Deployment

Training neural networks for neuromorphic deployment is non-trivial. There have been a variety of approaches proposed to adapt back-propagation or back-propagation-like algorithms appropriate for training. Considering that these networks often have very different performance characteristics than tra...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Parsa, Maryam, Schuman, Catherine D, Date, Prasanna, Rose, Derek C, Kay, Bill, J Parker Mitchell, Young, Steven R, Dellana, Ryan, Severa, William, Potok, Thomas E, Kaushik, Roy
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 21.04.2020
Subjects
Algorithms
Back propagation
Bayesian analysis
Communication
Communication networks
Communications networks
Datasets
Hardware
Network topologies
Neural networks
Optimization
Training
Online AccessGet full text

Cover

More Information
Summary:Training neural networks for neuromorphic deployment is non-trivial. There have been a variety of approaches proposed to adapt back-propagation or back-propagation-like algorithms appropriate for training. Considering that these networks often have very different performance characteristics than traditional neural networks, it is often unclear how to set either the network topology or the hyperparameters to achieve optimal performance. In this work, we introduce a Bayesian approach for optimizing the hyperparameters of an algorithm for training binary communication networks that can be deployed to neuromorphic hardware. We show that by optimizing the hyperparameters on this algorithm for each dataset, we can achieve improvements in accuracy over the previous state-of-the-art for this algorithm on each dataset (by up to 15 percent). This jump in performance continues to emphasize the potential when converting traditional neural networks to binary communication applicable to neuromorphic hardware.
ISSN:2331-8422