IMBUE: In-Memory Boolean-to-CUrrent Inference ArchitecturE for Tsetlin Machines

Ghazal, O; Singh, S; Rahman, T; Yu, S; Zheng, Y; Balsamo, D; Patkar, S; Merchant, F; Xia, F; Yakovlev, A; Shafik, R

doi:10.1109/ISLPED58423.2023.10244315

IMBUE: In-Memory Boolean-to-CUrrent Inference ArchitecturE for Tsetlin Machines

Lookup NU author(s): Dr Omar Awf ORCiD, Tousif Rahman ORCiD, Dr Shengqi Yu, Yujin Zheng ORCiD, Dr Domenico Balsamo ORCiD, Dr Farhad Merchant, Dr Fei Xia, Professor Alex Yakovlev ORCiD, Professor Rishad Shafik ORCiD

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

In-memory computing for Machine Learning (ML) applications remedies the von Neumann bottlenecks by organizing computation to exploit parallelism and locality. Non-volatile memory devices such as Resistive RAM (ReRAM) offer integrated switching and storage capabilities showing promising performance for ML applications. However, ReRAM devices have design challenges, such as nonlinear digital-analog conversion and circuit overheads. This paper proposes an In-Memory Boolean-to-Current Inference Architecture (IMBUE) that uses ReRAM-transistor cells to eliminate the need for such conversions. IMBUE processes Boolean feature inputs expressed as digital voltages and generates parallel current paths based on resistive memory states. The proportional column current is then translated back to the Boolean domain for further digital processing. The IMBUE architecture is inspired by the Tsetlin Machine (TM), an emerging ML algorithm based on intrinsically Boolean logic. The IMBUE architecture demonstrates significant performance improvements over binarized convolutional neural networks and digital TM in-memory implementations, achieving up to a 12.99x and 5.28x increase, respectively.