BNN - hardware versus software

https://www.slideserve.com/darleneandre/deep-neural-network-optimization-binary-neural-networks-powerpoint-ppt-presentation

Software approaches

algorithms

• 2016 Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1 Matthieu Courbariaux

Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations Itay Hubara, Matthieu Courbariaux

• 2020 Controlling Information Capacity of Binary Neural Network Dmitry Ignatov, Andrey Ignatov

Hardware approaches

2018 XNOR Neural Engine: a Hardware Accelerator IP for 21.6 fJ/op Binary Neural Network Inference

+++ 2018 Stanford An Always-On 3.8 μJ/86% CIFAR-10 Mixed-Signal Binary CNN Processor With All Memory on Chip in 28-nm CMOS Daniel Bankman

2020 - Review An Updated Survey of Efficient Hardware Architectures for Accelerating Deep Convolutional Neural Networks Maurizio Capra

https://larq.dev/

video tinyML Talks Lukas Geiger: Binarized Neural Networks on microcontrollers

• short intro about Plumerai

MLSys 2021: Design, Benchmark, and Deploy Binarized Neural Networks with Larq Compute Engine

more general

2015 Memristor-Based Multilayer Neural Networks With Online Gradient Descent Training