Waves and symbols in neuromorphic hardware: from analog signal processing to digital computing on the same computational substrate

2025-02-27

AI Summary

Conventional neuromorphic systems suffer from a fundamental modality divide between analog signal processing and digital symbolic computation. Method: This work proposes a cortical-microcircuit-inspired reconfigurable spiking neural network architecture, theoretically characterizing and experimentally validating a controllable phase transition mechanism in hybrid-feedback spiking networks between analog and digital modalities. It integrates soft winner-take-all circuits, dynamically reconfigurable analog neurons and synapses, and mixed-signal brain-inspired chips. Contribution/Results: The architecture enables millisecond-scale, noise-robust, seamless modality switching on a single hardware substrate, supporting concurrent analog filtering, pattern recognition, and digital logic operations within shared computational resources. This breaks the traditional modality isolation barrier in neuromorphic systems and establishes a novel paradigm for general-purpose brain-inspired computing.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Hardware-aware ML
• Computer Vision: Multi-modal Vision

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Neural systems use the same underlying computational substrate to carry out analog filtering and signal processing operations, as well as discrete symbol manipulation and digital computation. Inspired by the computational principles of canonical cortical microcircuits, we propose a framework for using recurrent spiking neural networks to seamlessly and robustly switch between analog signal processing and categorical and discrete computation. We provide theoretical analysis and practical neural network design tools to formally determine the conditions for inducing this switch. We demonstrate the robustness of this framework experimentally with hardware soft Winner-Take-All and mixed-feedback recurrent spiking neural networks, implemented by appropriately configuring the analog neuron and synapse circuits of a mixed-signal neuromorphic processor chip.

Problem

• Integrate analog and digital computations
• Use spiking neural networks
• Develop neuromorphic hardware framework

Innovation

• Recurrent spiking neural networks
• Analog and digital computation
• Neuromorphic processor chip