Nonlinearity Modeling for Mixed-Signal Inference Accelerators in Training Frameworks

Conrad J, Jiang B, Kaesser P, Ortmanns M, Belagiannis V (2021)

Publication Type: Conference contribution

Publication year: 2021

Publisher: Institute of Electrical and Electronics Engineers Inc.

Conference Proceedings Title: 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings

Event location: Dubai, ARE

ISBN: 9781728182810

DOI: 10.1109/ICECS53924.2021.9665503

Abstract

In this work, an approach for modeling the nonlinearity of a mixed-signal neural-network accelerator in training frameworks is presented. We extend the state-of-the-art by modeling a mixed-signal neuron in a neural-network training framework such as TensorFlow. It is shown how the nonlinearity can be integrated in the anyhow required quantizer models. The parameters of the nonlinearity model of a single neuron are found by a preliminary training, where the model variables are treated as learnable parameters, while the behavior of the modeled neuron is fitted to circuit-simulation or -test data. The model is never moved to another toolchain and the entire model extraction process and the process of training a neural network under the influence of circuit-nonlinearities happen in the training framework, where TensorFlow is chosen for this work. We evaluate the approach by analyzing how a full-scale VGG-16 based CIFAR-10 classifier adapts a known neuron nonlinearity. The impact of the nonlinearities can be removed by training and without performing improvements on circuit level.

Authors with CRIS profile

Vasileios Belagiannis

Involved external institutions

Universität Ulm DE Germany (DE)

How to cite

APA:

Conrad, J., Jiang, B., Kaesser, P., Ortmanns, M., & Belagiannis, V. (2021). Nonlinearity Modeling for Mixed-Signal Inference Accelerators in Training Frameworks. In 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings. Dubai, ARE: Institute of Electrical and Electronics Engineers Inc..

MLA:

Conrad, Joschua, et al. "Nonlinearity Modeling for Mixed-Signal Inference Accelerators in Training Frameworks." Proceedings of the 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021, Dubai, ARE Institute of Electrical and Electronics Engineers Inc., 2021.

BibTeX: Download