Efficient Neural Network Implementation with Quadratic Neuron

Zirui Xu; Jinjun Xiong; Fuxun Yu; Xiang Chen

Efficient Neural Network Implementation with Quadratic Neuron

Networking and Internet Architecture 2020-11-24 v1

Authors: Zirui Xu , Jinjun Xiong , Fuxun Yu , Xiang Chen

Abstract

Previous works proved that the combination of the linear neuron network with nonlinear activation functions (e.g. ReLu) can achieve nonlinear function approximation. However, simply widening or deepening the network structure will introduce some training problems. In this work, we are aiming to build a comprehensive second-order CNN implementation framework that includes neuron/network design and system deployment optimization.

Keywords

activation functions neural networks neural network

Cite

@article{arxiv.2011.10813,
  title  = {Efficient Neural Network Implementation with Quadratic Neuron},
  author = {Zirui Xu and Jinjun Xiong and Fuxun Yu and Xiang Chen},
  journal= {arXiv preprint arXiv:2011.10813},
  year   = {2020}
}

Comments

2 pages, 1 figure, accepted by 2020 3rd IBM IEEE CAS/EDS AI Compute Symposium

Related papers

View all related →