AUTHOR=Spoon Katie , Tsai Hsinyu , Chen An , Rasch Malte J. , Ambrogio Stefano , Mackin Charles , Fasoli Andrea , Friz Alexander M. , Narayanan Pritish , Stanisavljevic Milos , Burr Geoffrey W. TITLE=Toward Software-Equivalent Accuracy on Transformer-Based Deep Neural Networks With Analog Memory Devices JOURNAL=Frontiers in Computational Neuroscience VOLUME=15 YEAR=2021 URL=https://www.frontiersin.org/journals/computational-neuroscience/articles/10.3389/fncom.2021.675741 DOI=10.3389/fncom.2021.675741 ISSN=1662-5188 ABSTRACT=

Recent advances in deep learning have been driven by ever-increasing model sizes, with networks growing to millions or even billions of parameters. Such enormous models call for fast and energy-efficient hardware accelerators. We study the potential of Analog AI accelerators based on Non-Volatile Memory, in particular Phase Change Memory (PCM), for software-equivalent accurate inference of natural language processing applications. We demonstrate a path to software-equivalent accuracy for the GLUE benchmark on BERT (Bidirectional Encoder Representations from Transformers), by combining noise-aware training to combat inherent PCM drift and noise sources, together with reduced-precision digital attention-block computation down to INT6.