Phase Change Memory-based Hardware Accelerators for Deep Neural Networks (invited)

Geoffrey W. Burr, Pritish Narayanan, Stefano Ambrogio, Atsuya Okazaki, Hsinyu Tsai, Kohji Hosokawa, Charles Mackin, Akiyo Nomura, Takeo Yasuda, J. Demarest, Kevin Brew, Victor Chan, Samuel Choi, T. Gordon, T. M. Levin, Alexander M. Friz, Masatoshi Ishii, Yasuteru Kohda, An Chen, Andrea Fasoli et al. (5 additional authors not shown)

Abstract: Analog non-volatile memory (NVM)-based accelerators for deep neural networks implement multiply-accumulate (MAC) operations – in parallel, on large arrays of resistive devices – by using Ohm’s law and Kirchhoff’s current law. By completely avoiding weight motion, such fully weight-stationary systems can offer a unique combination of low latency, high throughput, and high energy-efficiency (e.g., high TeraOPS/W). Yet since most Deep Neural Networks (DNNs) require only modest (e.g., 4-bit) precision in synaptic operations, such systems can still deliver “software-equivalent” accuracies on a wide range of models. We describe a 14-nm inference chip, comprising multiple 512×512 arrays of Phase Change Memory (PCM) devices, which can deliver software-equivalent inference accuracy for MNIST handwritten-digit recognition and recurrent LSTM benchmarks, and discuss various PCM challenges such as conductance drift and noise.