# A Spiking Network for Inference of Relations Trained with Neuromorphic Backpropagation

## TL;DR

This paper introduces a novel multi-layer spiking neural network trained with a neuromorphic backpropagation algorithm, capable of relational inference on complex stimuli with on-chip learning, suitable for autonomous sensor systems.

## Contribution

It presents the first spiking neural network architecture with on-chip learning for relational inference on complex visual data.

## Key findings

- Achieves better performance with fewer neurons than previous biologically-inspired models.
- Successfully performs relational inference on arithmetic and visual XOR tasks.
- Demonstrates feasibility of on-chip learning in neuromorphic hardware for complex stimuli.

## Abstract

The increasing need for intelligent sensors in a wide range of everyday objects requires the existence of low power information processing systems which can operate autonomously in their environment. In particular, merging and processing the outputs of different sensors efficiently is a necessary requirement for mobile agents with cognitive abilities. In this work, we present a multi-layer spiking neural network for inference of relations between stimuli patterns in dedicated neuromorphic systems. The system is trained with a new version of the backpropagation algorithm adapted to on-chip learning in neuromorphic hardware: Error gradients are encoded as spike signals which are propagated through symmetric synapses, using the same integrate-and-fire hardware infrastructure as used during forward propagation. We demonstrate the strength of the approach on an arithmetic relation inference task and on visual XOR on the MNIST dataset. Compared to previous, biologically-inspired implementations of networks for learning and inference of relations, our approach is able to achieve better performance with less neurons. Our architecture is the first spiking neural network architecture with on-chip learning capabilities, which is able to perform relational inference on complex visual stimuli. These features make our system interesting for sensor fusion applications and embedded learning in autonomous neuromorphic agents.