A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers
21 Mar 2017 Oak Ridge - Researchers from Oak Ridge, the University of Tenessee and USC published "A Study of Complex Deep Learning Networks on High Performance, Neuromorphic, and Quantum Computers" . Current Deep Learning approaches have been very successful using convolutional neural networks (CNN) trained on large graphical processing units (GPU)-based computers. The authors see three limitations of this approach are: 1. they are based on a simple layered network topology, i.e., highly connected layers, without intra-layer connections; 2. the networks are manually configured to achieve optimal results, and 3. the implementation of neuron model is expensive in both cost and power. In a paper published on Arvid the authors, evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing (HPC) to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation.
The authors, Thomas Potok, Catherine Schuman, Steven Young, Robert Patton, Federico Spedalieri, Jeremy Liu, Ke-Thia Yao, Garrett Rose, Gangotree Chakma used the MNIST dataset for their experiment, due to input size limitations of current quantum computers. Their results show the feasibility of using the three architectures in tandem to address the above deep learning limitations.
The paper shows a quantum computer can find high quality values of intra-layer connections weights, in a tractable time as the complexity of the network increases; an HPC system can find optimal layer-based topologies; and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware.