# Neuromorphic Architecture Optimization for Task-Specific Dynamic Learning

## TL;DR

This paper presents a framework for optimizing neuromorphic architectures with task-specific learning rules using meta-learning and hyperparameter tuning, demonstrated on MNIST datasets with insect-inspired networks.

## Contribution

It introduces a novel meta-learning optimization approach for neuromorphic systems, highlighting dataset-dependent optimal learning rules and hyperparameters.

## Key findings

- Optimal learning rules vary with datasets.
- Meta-learning improves task-specific performance.
- Insect-inspired networks benefit from flexible learning rules.

## Abstract

The ability to learn and adapt in real time is a central feature of biological systems. Neuromorphic architectures demonstrating such versatility can greatly enhance our ability to efficiently process information at the edge. A key challenge, however, is to understand which learning rules are best suited for specific tasks and how the relevant hyperparameters can be fine-tuned. In this work, we introduce a conceptual framework in which the learning process is integrated into the network itself. This allows us to cast meta-learning as a mathematical optimization problem. We employ DeepHyper, a scalable, asynchronous model-based search, to simultaneously optimize the choice of meta-learning rules and their hyperparameters. We demonstrate our approach with two different datasets, MNIST and FashionMNIST, using a network architecture inspired by the learning center of the insect brain. Our results show that optimal learning rules can be dataset-dependent even within similar tasks. This dependency demonstrates the importance of introducing versatility and flexibility in the learning algorithms. It also illuminates experimental findings in insect neuroscience that have shown a heterogeneity of learning rules within the insect mushroom body.