Deep Learning in Memristive Nanowire Networks
This work addresses hardware inefficiencies for sparse neural networks in AI/ML, though it is incremental as it builds on existing memristive nanowire concepts.
The authors tackled the inefficiency of analog crossbar architectures for large sparse neural network layers by proposing the MN3 (Memristive Nanowire Neural Network) as a hardware architecture, demonstrating through simulations that it can function as a deep neural network and achieve a 1.61% error rate on the MNIST dataset, comparable to software-based networks.
Analog crossbar architectures for accelerating neural network training and inference have made tremendous progress over the past several years. These architectures are ideal for dense layers with fewer than roughly a thousand neurons. However, for large sparse layers, crossbar architectures are highly inefficient. A new hardware architecture, dubbed the MN3 (Memristive Nanowire Neural Network), was recently described as an efficient architecture for simulating very wide, sparse neural network layers, on the order of millions of neurons per layer. The MN3 utilizes a high-density memristive nanowire mesh to efficiently connect large numbers of silicon neurons with modifiable weights. Here, in order to explore the MN3's ability to function as a deep neural network, we describe one algorithm for training deep MN3 models and benchmark simulations of the architecture on two deep learning tasks. We utilize a simple piecewise linear memristor model, since we seek to demonstrate that training is, in principle, possible for randomized nanowire architectures. In future work, we intend on utilizing more realistic memristor models, and we will adapt the presented algorithm appropriately. We show that the MN3 is capable of performing composition, gradient propagation, and weight updates, which together allow it to function as a deep neural network. We show that a simulated multilayer perceptron (MLP), built from MN3 networks, can obtain a 1.61% error rate on the popular MNIST dataset, comparable to equivalently sized software-based network. This work represents, to the authors' knowledge, the first randomized nanowire architecture capable of reproducing the backpropagation algorithm.