EVE: Environmental Adaptive Neural Network Models for Low-power Energy Harvesting System

TL;DR

EVE introduces an autoML framework for designing energy-efficient neural networks with shared weights, optimized for low-power energy harvesting IoT devices facing environmental energy variability.

Contribution

The paper presents a novel autoML co-exploration framework that creates shared-weight neural models with adjustable sparsity, enabling adaptive, low-power IoT applications powered by ambient energy.

Findings

Models are 2.5X faster than baseline models.

Shared weights reduce memory footprint significantly.

Framework adapts models to environmental energy changes.

Abstract

IoT devices are increasingly being implemented with neural network models to enable smart applications. Energy harvesting (EH) technology that harvests energy from ambient environment is a promising alternative to batteries for powering those devices due to the low maintenance cost and wide availability of the energy sources. However, the power provided by the energy harvester is low and has an intrinsic drawback of instability since it varies with the ambient environment. This paper proposes EVE, an automated machine learning (autoML) co-exploration framework to search for desired multi-models with shared weights for energy harvesting IoT devices. Those shared models incur significantly reduced memory footprint with different levels of model sparsity, latency, and accuracy to adapt to the environmental changes. An efficient on-device implementation architecture is further developed to efficiently execute each model on device. A run-time model extraction algorithm is proposed that retrieves individual model with negligible overhead when a specific model mode is triggered.Experimental results show that the neural networks models generated by EVE is on average 2.5X times faster than the baseline models without pruning and shared weights.