Synaptic bundle theory for spike-driven sensor-motor system: More than eight independent synaptic bundles collapse reward-STDP learning

2025-08-20

AI Summary

This work addresses the instability of reward-modulated spike-timing-dependent plasticity (STDP) in spiking sensorimotor systems. To mitigate learning collapse, we propose and validate the “synaptic bundle” theory: by constructing spiking neural circuits with tunable numbers of independent synaptic bundles, we identify a critical capacity threshold—learning fails when the number of independent bundles exceeds eight due to misaligned weight update directions; conversely, with ≤8 bundles, reducing motor neuron count accelerates convergence. Quantitative experiments characterize the trade-offs among synaptic bundle count, motor neuron population size, learning success rate, and convergence speed. This study establishes, for the first time, functional boundaries of spiking signals in motor control, providing essential parametric guidelines and theoretical foundations for designing robust spiking intelligent systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Neuro-Symbolic Learning

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: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Neuronal spikes directly drive muscles and endow animals with agile movements, but applying the spike-based control signals to actuators in artificial sensor-motor systems inevitably causes a collapse of learning. We developed a system that can vary emph{the number of independent synaptic bundles} in sensor-to-motor connections. This paper demonstrates the following four findings: (i) Learning collapses once the number of motor neurons or the number of independent synaptic bundles exceeds a critical limit. (ii) The probability of learning failure is increased by a smaller number of motor neurons, while (iii) if learning succeeds, a smaller number of motor neurons leads to faster learning. (iv) The number of weight updates that move in the opposite direction of the optimal weight can quantitatively explain these results. The functions of spikes remain largely unknown. Identifying the parameter range in which learning systems using spikes can be constructed will make it possible to study the functions of spikes that were previously inaccessible due to the difficulty of learning.

Problem

• Spike-based control causes artificial system learning collapse
• Identifies critical limit of motor neurons for stable learning
• Quantifies weight update errors in spike-driven reward learning

Innovation

• Spike-driven system with independent synaptic bundles
• Collapse prevention via critical bundle limit control
• Opposite-direction weight updates explain learning dynamics