A 65-nm Neuromorphic Image Classification Processor With Energy-Efficient Training Through Direct Spike-Only Feedback

• Published in: IEEE journal of solid-state circuits Vol. 55; no. 1; pp. 108 - 119
• Main Authors: Park, Jeongwoo, Lee, Juyun, Jeon, Dongsuk
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerators; Algorithms; Artificial intelligence; Artificial neural networks; Back propagation; Cognitive tasks; Computational efficiency; Design modifications; digital integrated circuits; Edge computing; Electronic devices; Energy; Energy consumption; Handwriting; Hardware; Image classification; Inference; learning systems; Machine learning; Machine learning algorithms; Microprocessors; multi layer perceptrons; Neural networks; Neuromorphics; Neurons; Task analysis; Training; very large-scale integration
• ISSN: 0018-9200; 1558-173X
• DOI: 10.1109/JSSC.2019.2942367

Recent advances in neural network (NN) and machine learning algorithms have sparked a wide array of research in specialized hardware, ranging from high-performance NN accelerators for use inside the server systems to energy-efficient edge computing systems. While most of these studies have focused on designing inference engines, implementing the training process of an NN for energy-constrained mobile devices has remained to be a challenge due to the requirement of higher numerical precision. In this article, we aim to build an on-chip learning system that would show highly energy-efficient training for NNs without degradation in the performance for machine learning tasks. To achieve this goal, we adapt and optimize a neuromorphic learning algorithm and propose hardware design techniques to fully exploit the properties of the modifications. We verify that our system achieves energy-efficient training with only 7.5% more energy consumption compared with its highly efficient inference of 236 nJ/image on the handwritten digit [Modified National Institute of Standards and Technology database (MNIST)] images. Moreover, our system achieves 97.83% classification accuracy on the MNIST test data set, which outperforms prior neuromorphic on-chip learning systems and is close to the performance of the conventional method for training deep neural networks (NNs), the backpropagation.