3 papers
cs.LG2018
Pre-Defined Sparse Neural Networks with Hardware Acceleration
Sourya Dey, Kuan-Wen Huang, Peter A. Beerel +1
Neural networks have proven to be extremely powerful tools for modern artificial intelligence applications, but computational and storage complexity remain limiting factors. This p…
cs.DC2018
A Highly Parallel FPGA Implementation of Sparse Neural Network Training
Sourya Dey, Diandian Chen, Zongyang Li +4
We demonstrate an FPGA implementation of a parallel and reconfigurable architecture for sparse neural networks, capable of on-chip training and inference. The network connectivity…
cs.LG2017
Characterizing Sparse Connectivity Patterns in Neural Networks
Sourya Dey, Kuan-Wen Huang, Peter A. Beerel +1
We propose a novel way of reducing the number of parameters in the storage-hungry fully connected layers of a neural network by using pre-defined sparsity, where the majority of co…