Abstract:SNNs are an active research domain towards energy efficient machine intelligence. Compared to conventional ANNs, SNNs use temporal spike data and bio-plausible neuronal activation functions such as Leaky-Integrate Fire/Integrate Fire (LIF/IF) for data processing. However, SNNs incur significant dot-product operations causing high memory and computation overhead in standard von-Neumann computing platforms. Today, In-Memory Computing (IMC) architectures have been proposed to alleviate the "memory-wall bottleneck" prevalent in von-Neumann architectures. Although recent works have proposed IMC-based SNN hardware accelerators, the following have been overlooked- 1) the adverse effects of crossbar non-ideality on SNN performance due to repeated analog dot-product operations over multiple time-steps, 2) hardware overheads of essential SNN-specific components such as the LIF/IF and data communication modules. To this end, we propose SpikeSim, a tool that can perform realistic performance, energy, latency and area evaluation of IMC-mapped SNNs. SpikeSim consists of a practical monolithic IMC architecture called SpikeFlow for mapping SNNs. Additionally, the non-ideality computation engine (NICE) and energy-latency-area (ELA) engine performs hardware-realistic evaluation of SpikeFlow-mapped SNNs. Based on 65nm CMOS implementation and experiments on CIFAR10, CIFAR100 and TinyImagenet datasets, we find that the LIF/IF neuronal module has significant area contribution (>11% of the total hardware area). We propose SNN topological modifications leading to 1.24x and 10x reduction in the neuronal module's area and the overall energy-delay-product value, respectively. Furthermore, in this work, we perform a holistic comparison between IMC implemented ANN and SNNs and conclude that lower number of time-steps are the key to achieve higher throughput and energy-efficiency for SNNs compared to 4-bit ANNs.

Sparsity-Aware In-Memory Neuromorphic Computing Unit With Configurable Topology of Hybrid Spiking and Artificial Neural Network

A A 22nm 0.43pJ/SOP Sparsity-Aware In-Memory Neuromorphic Computing System with Hybrid Spiking and Artificial Neural Network and Configurable Topology

30.2 A 22nm 0.26nW/Synapse Spike-Driven Spiking Neural Network Processing Unit Using Time-Step-First Dataflow and Sparsity-Adaptive In-Memory Computing

A Scatter-and-Gather Spiking Convolutional Neural Network on a Reconfigurable Neuromorphic Hardware

An 1.38nj/inference Clock-Free Mixed-Signal Neuromorphic Architecture Using ReL-PSP Function and Computing-in-Memory.

An Energy-Efficient Computing-in-Memory NN Processor with Set-Associate Blockwise Sparsity and Ping-Pong Weight Update

Multicore Spiking Neuromorphic Chip in 180-nm with ReRAM Synapses and Digital Neurons

55 Nm CMOS Mixed-Signal Neuromorphic Circuits for Constructing Energy-Efficient Reconfigurable SNNs

A Hybrid Precision Low Power Computing-in-memory Architecture for Neural Networks

A 64K-Neuron 64M-1b-Synapse 2.64pJ/SOP Neuromorphic Chip With All Memory on Chip for Spike-Based Models in 65nm CMOS

DS-CIM: A 40nm Asynchronous Dual-Spike Driven, MRAM Compute-In-Memory Macro for Spiking Neural Network

SpikeSim: An end-to-end Compute-in-Memory Hardware Evaluation Tool for Benchmarking Spiking Neural Networks

Sparsity-Aware Non-Volatile Computing-In-Memory Macro with Analog Switch Array and Low-Resolution Current-Mode ADC.

When In-memory Computing Meets Spiking Neural Networks -- A Perspective on Device-Circuit-System-and-Algorithm Co-design

Design and Implementation of a Charge-Sharing In-Memory-computing Macro with Sparse Feature for Quantized Neural Network

You Only Spike Once: Improving Energy-Efficient Neuromorphic Inference to ANN-Level Accuracy

NeuroNAS: A Framework for Energy-Efficient Neuromorphic Compute-in-Memory Systems using Hardware-Aware Spiking Neural Architecture Search

An Area- and Energy-Efficient Spiking Neural Network with Spike-Time-Dependent Plasticity Realized with SRAM Processing-in-Memory Macro and On-Chip Unsupervised Learning.

NBSSN: A Neuromorphic Binary Single-Spike Neural Network for Efficient Edge Intelligence.

Hardware/Software co-design with ADC-Less In-memory Computing Hardware for Spiking Neural Networks

Efficient Discrete Temporal Coding Spike-Driven In-Memory Computing Macro for Deep Neural Network Based on Nonvolatile Memory.