ReckOn: A 28nm Sub-mm2 Task-Agnostic Spiking Recurrent Neural Network Processor Enabling On-Chip Learning over Second-Long Timescales

Abstract

A robust real-world deployment of autonomous edge devices requires on-chip adaptation to user-, environment- and task-induced variability. Due to on-chip memory constraints, prior learning devices were limited to static stimuli with no temporal contents. We propose a 0.45-mm${}^2$ spiking RNN processor enabling task-agnostic online learning over seconds, which we demonstrate for navigation, gesture recognition, and keyword spotting within a 0.8-% memory overhead and a <150-$\mu$W training power budget.

arXiv commentPublished in the 2022 IEEE International Solid-State Circuits Conference (ISSCC), 2022

References in corpus (1)

• The Heidelberg spiking datasets for the systematic evaluation of spiking neural networks

Cited by in corpus (15)

• Bottom-up and top-down approaches for the design of neuromorphic processing systems: Tradeoffs and synergies between natural and artificial intelligence
• Neuromorphic Intermediate Representation: A Unified Instruction Set for Interoperable Brain-Inspired Computing
• Spiker+: a framework for the generation of efficient Spiking Neural Networks FPGA accelerators for inference at the edge
• Advancing Spatio-Temporal Processing in Spiking Neural Networks through Adaptation
• ETLP: Event-based Three-factor Local Plasticity for online learning with neuromorphic hardware
• Demonstrating Analog Inference on the BrainScaleS-2 Mobile System
• DeltaKWS: A 65nm 36nJ/Decision Bio-inspired Temporal-Sparsity-Aware Digital Keyword Spotting IC with 0.6V Near-Threshold SRAM
• Continual Learning with Neuromorphic Computing: Foundations, Methods, and Emerging Applications
• Compressed Latent Replays for Lightweight Continual Learning on Spiking Neural Networks
• Benchmarking Spiking Neural Network Learning Methods with Varying Locality
• A Systematic Review of ECG Arrhythmia Classification: Adherence to Standards, Fair Evaluation, and Embedded Feasibility
• Heterogeneous SoC Integrating an Open-Source Recurrent SNN Accelerator for Neuromorphic Edge Computing on FPGA
• TESS: A Scalable Temporally and Spatially Local Learning Rule for Spiking Neural Networks
• Threshold Modulation for Online Test-Time Adaptation of Spiking Neural Networks
• Three factor delay learning rules for spiking neural networks