Stochastic Quantum Spiking Neural Networks with Quantum Memory and Local Learning

2025-06-26

AI Summary

Existing quantum spiking neurons rely on single-qubit classical memory, require repeated measurements to estimate spiking probabilities, and depend on classical backpropagation for training—limiting hardware deployability. Method: We propose the first random spiking neuron with multi-qubit embedded quantum memory, enabling event-driven spike generation from a single measurement; design a quantum-friendly learning rule leveraging only local synaptic information, eliminating global backpropagation; and construct a modular, scalable, fully quantum-trainable spiking network architecture. Contribution/Results: Our approach preserves brain-inspired temporal processing efficiency while harnessing the exponential capacity of quantum state space to enhance representational power. It enables end-to-end quantum-native training and inference, offering a novel paradigm for neuromorphic computing tailored to near-term quantum hardware.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Neuromorphic and quantum computing have recently emerged as promising paradigms for advancing artificial intelligence, each offering complementary strengths. Neuromorphic systems built on spiking neurons excel at processing time-series data efficiently through sparse, event-driven computation, consuming energy only upon input events. Quantum computing, on the other hand, leverages superposition and entanglement to explore feature spaces that are exponentially large in the number of qubits. Hybrid approaches combining these paradigms have begun to show potential, but existing quantum spiking models have important limitations. Notably, prior quantum spiking neuron implementations rely on classical memory mechanisms on single qubits, requiring repeated measurements to estimate firing probabilities, and they use conventional backpropagation on classical simulators for training. Here we propose a stochastic quantum spiking (SQS) neuron model that addresses these challenges. The SQS neuron uses multi-qubit quantum circuits to realize a spiking unit with internal quantum memory, enabling event-driven probabilistic spike generation in a single shot. Furthermore, we outline how networks of SQS neurons -- dubbed SQS neural networks (SQSNNs) -- can be trained via a hardware-friendly local learning rule, eliminating the need for global classical backpropagation. The proposed SQSNN model fuses the time-series efficiency of neuromorphic computing with the exponentially large inner state space of quantum computing, paving the way for quantum spiking neural networks that are modular, scalable, and trainable on quantum hardware.

Problem

• Develop quantum spiking neuron with internal quantum memory
• Enable single-shot event-driven probabilistic spike generation
• Train networks without global classical backpropagation

Innovation

• Multi-qubit circuits enable quantum memory spiking
• Local learning rule replaces global backpropagation
• Event-driven probabilistic spikes in single shot