Highly Optimized Hardware Morphological Neural Network Through Stochastic Computing and Tropical Pruning

• Published in: IEEE journal on emerging and selected topics in circuits and systems 2023-03, Vol.13 (1), p.249-256
• Main Authors: Rossello, Josep L., Font-Rossello, Joan, Frasser, Christiam F., Moran, Alejandro, Skibinsky-Gitlin, Erik Sebastian, Canals, Vincent, Roca, Miquel
• Format: Article
• Language: English
• Subjects: Algebra; approximate computing computers; Artificial neural networks (ANNs); Biological neural networks; Computation; computational and artificial intelligence; Energy dissipation; Hardware; hardware acceleration computers; information processing; Logic gates; Multi-layer neural network; neural network hardware computers; Neural networks; Neurons; probabilistic computing; Symbols
• ISSN: 2156-3357; 2156-3365
• DOI: 10.1109/JETCAS.2022.3226292

This work aimed to enhance a previous neural network hardware implementation based on an efficient combination of Stochastic Computing (SC) and Morphological Neural Networks (MNN). This enhancement focused on exploiting the natural ease of morphological neurons to be pruned in order to drastically shrink the hardware resources and increase the compactness of our network. That is why we extended our original hybrid two-layer neural network to classify the MNIST problem, a much more demanding benchmark with about 160,000 trainable parameters. The 92% of the weights of the morphological layer were discarded, allowing a drastic shrinkage of the hardware resources and power dissipation without degrading test accuracy. An extensive comparison with other recent neural networks designs shows that the proposed design achieves significant improvements in terms of energy efficiency, throughput and hardware resources.