## BNN algorithms

2019 [Back to Simplicity: How to Train Accurate BNNs from Scratch?](https://arxiv.org/abs/1906.08637) Joseph Bethge


## non-idealities and noise


[Committee machines—a universal method to deal with non-idealities in memristor-based neural networks](https://www.nature.com/articles/s41467-020-18098-0) Joksas, A. J. Kenyon & A. Mehonic

2019 [Stochastic Computing for Hardware Implementation of Binarized Neural Networks](https://arxiv.org/abs/1906.00915)

> We evidence that the training procedure should be adapted for use with stochastic computing. Finally, the ASIC implementation of our scheme is investigated, in a system that closely associates logic and memory, implemented by Spin Torque Magnetoresistive Random Access Memory. This analysis shows that the **stochastic computing approach can allow considerable savings** with regards to conventional Binarized Neural networks in terms of area (62% area reduction on the Fashion-MNIST task)

- [Improving Noise Tolerance of Mixed-Signal Neural Networks](https://arxiv.org/abs/1904.01705) Michael Klachko, Mohammad Reza Mahmoodi, Dmitri B. Strukov

> We apply various methods to improve noise robustness of the network and demonstrate an effective way to optimize useful
> signal ranges through adaptive signal clipping


## training

2020 [Accurate deep neural network inference using computational phase-change memory](https://www.nature.com/articles/s41467-020-16108-9) Vinay Joshi

> ....However, due to device variability and noise, the network needs to be trained in a specific way so that **transferring the digitally trained weights to the analog resistive memory devices will not result in significant loss of accuracy**. Here, we introduce a methodology to train ResNet-type convolutional neural networks that results in no appreciable accuracy loss when transferring weights to phase-change memory (PCM) devices. We also propose a compensation technique that exploits the batch normalization parameters