Print Email Facebook Twitter Energy-Efficient SNN Implementation Using RRAM-Based Computation In-Memory (CIM) Title Energy-Efficient SNN Implementation Using RRAM-Based Computation In-Memory (CIM) Author El Arrassi, A.E. (Abdelmalek Essaadi University) Gebregiorgis, A.B. (TU Delft Computer Engineering) Haddadi, Anass El (Abdelmalek Essaadi University) Hamdioui, S. (TU Delft Quantum & Computer Engineering) Department Quantum & Computer Engineering Date 2022 Abstract Spiking Neural Networks (SNNs) can drastically improve the energy efficiency of neuromorphic computing through network sparsity and event-driven execution. Thus, SNNs have the potential to support practical cognitive tasks on resource constrained platforms, such as edge devices. To realize this, SNN requires energy-efficient hardware which can run applications with a limited energy budget. However, the conventional CMOS implementations cannot achieve this goal due to the various architectural and technological challenges. In this work, we address these issues by developing an energy-efficient and accurate SNN hardware based on Computation In-Memory (CIM) architecture using Resistive Random Access Memory (RRAM) devices. The developed SNN architecture is based on unsupervised Spike Time Dependent Plasticity (STDP) learning algorithm with online learning capability. Simulation results show that the proposed architecture is energy-efficient with a consumption of ≈20 fJ per spike, while maintaining state-of-the-art inference accuracy of 95% when evaluated using the MNIST dataset. Subject SNNRRAMIn-Memory ComputingSTDP To reference this document use: http://resolver.tudelft.nl/uuid:8fc5e807-1ccd-4593-b017-5c7847ae0cbf DOI https://doi.org/10.1109/VLSI-SoC54400.2022.9939654 Publisher IEEE Embargo date 2023-07-01 ISBN 978-1-6654-9006-1 Source Proceedings of the 2022 IFIP/IEEE 30th International Conference on Very Large Scale Integration (VLSI-SoC) Event 2022 IFIP/IEEE 30th International Conference on Very Large Scale Integration (VLSI-SoC), 2022-10-03 → 2022-10-05, Patras, Greece Bibliographical note Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. Part of collection Institutional Repository Document type conference paper Rights © 2022 A.E. El Arrassi, A.B. Gebregiorgis, Anass El Haddadi, S. Hamdioui Files PDF Energy_Efficient_SNN_Impl ... ry_CIM.pdf 3.1 MB Close viewer /islandora/object/uuid:8fc5e807-1ccd-4593-b017-5c7847ae0cbf/datastream/OBJ/view