Early Exiting Predictive Coding Neural Networks for Edge AI

TL;DR

This paper introduces a shallow predictive coding neural network with early exiting, inspired by brain efficiency, to enable resource-efficient, high-accuracy edge AI suitable for IoT devices.

Contribution

The authors propose a novel shallow bidirectional predictive coding network with early exiting, reducing computation and memory while maintaining accuracy for edge AI applications.

Findings

Achieves comparable accuracy to deep networks on CIFAR-10

Reduces computational complexity and memory footprint

Demonstrates biologically inspired architecture benefits for edge AI

Abstract

The Internet of Things is transforming various fields, with sensors increasingly embedded in wearables, smart buildings, and connected equipment. While deep learning enables valuable insights from IoT data, conventional models are too computationally demanding for resource-limited edge devices. Moreover, privacy concerns and real-time processing needs make local computation a necessity over cloud-based solutions. Inspired by the brain's energy efficiency, we propose a shallow bidirectional predictive coding network with early exiting, dynamically halting computations once a performance threshold is met. This reduces the memory footprint and computational overhead while maintaining high accuracy. We validate our approach using the CIFAR-10 dataset. Our model achieves performance comparable to deep networks with significantly fewer parameters and lower computational complexity, demonstrating the potential of biologically inspired architectures for efficient edge AI.