The Community for Technology Leaders
Green Image
<p>EPSILON, a large, working, VLSI device, demonstrates pulse stream methods in the wider context of analog neural networks. EPSILON uses dynamic weight storage techniques, but a nonvolatile alternative is desirable. To that end, we have developed an amorphous silicon memory, which we present in experiments incorporating the device in a modest pulse stream neural chip. We have also developed a target-based training algorithm, which we demonstrate in a prototype learning device using a realistic problem. Finally, we explore system-level problems in experiments with a second version of EPSILON in a small, autonomous robot.</p>
Alan F. Murray, Stephen Churcher, Alister Hamilton, Andrew J. Homes, Geoff B. Jackson, H. Martin Reekie, Robin J. Woodburn, "Pulse Stream VLSI Neural Networks", IEEE Micro, vol. 14, no. , pp. 29-39, June 1994, doi:10.1109/40.285222
83 ms
(Ver 3.3 (11022016))