9 October 1998 In-the-loop training algorithm for neural network implementation with digital weights
Author Affiliations +
Proceedings Volume 3517, Intelligent Systems in Design and Manufacturing; (1998) https://doi.org/10.1117/12.326913
Event: Photonics East (ISAM, VVDC, IEMB), 1998, Boston, MA, United States
In this paper, we propose a training algorithm for VLSI neural networks with digital weights and analog neurons using in-the-loop training strategy. The use of digital weights in a neural network implementation imposes new issues that are not present in simulation environments. One of the problems is that a neural network implementation will not work properly when using the digitized version of the continuous weight solution. This phenomenon is especially evident when the digital weight resolution is very low due to some fabrication constraints. In this paper the training strategies for dealing with digital weights are investigated. The proposed training algorithm is by measuring the sensitivity of each weight to its error function and then by perturbing the weights of higher sensitivity values to perform retraining process. Our experimental results indicate that the algorithm is feasible and particularly suitable for the digital weights with low number of bits.
© (1998) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jinming Yang, Jinming Yang, Graham A. Jullien, Graham A. Jullien, Majid A. Ahmadi, Majid A. Ahmadi, W. C. Miller, W. C. Miller, } "In-the-loop training algorithm for neural network implementation with digital weights", Proc. SPIE 3517, Intelligent Systems in Design and Manufacturing, (9 October 1998); doi: 10.1117/12.326913; https://doi.org/10.1117/12.326913

Back to Top