DTNN: Energy-efficient Inference with Dendrite Tree Inspired Neural Networks for Edge Vision Applications

TL;DR

This paper introduces DTNN, a dendrite tree inspired neural network architecture that significantly reduces energy consumption during inference on edge devices by eliminating costly operations through activation quantization and table lookup, maintaining accuracy.

Contribution

The paper proposes a novel DTNN architecture that enables energy-efficient inference by removing weight access and arithmetic computations, validated across multiple models and datasets.

Findings

Achieved up to 64.9X energy savings on VGG-11 with ImageNet.

Maintained negligible accuracy loss across various models and datasets.

Demonstrated superior energy efficiency and latency on FPGA implementations.

Abstract

Deep neural networks (DNN) have achieved remarkable success in computer vision (CV). However, training and inference of DNN models are both memory and computation intensive, incurring significant overhead in terms of energy consumption and silicon area. In particular, inference is much more cost-sensitive than training because training can be done offline with powerful platforms, while inference may have to be done on battery powered devices with constrained form factors, especially for mobile or edge vision applications. In order to accelerate DNN inference, model quantization was proposed. However previous works only focus on the quantization rate without considering the efficiency of operations. In this paper, we propose Dendrite-Tree based Neural Network (DTNN) for energy-efficient inference with table lookup operations enabled by activation quantization. In DTNN both costly weight access and arithmetic computations are eliminated for inference. We conducted experiments on various kinds of DNN models such as LeNet-5, MobileNet, VGG, and ResNet with different datasets, including MNIST, Cifar10/Cifar100, SVHN, and ImageNet. DTNN achieved significant energy saving (19.4X and 64.9X improvement on ResNet-18 and VGG-11 with ImageNet, respectively) with negligible loss of accuracy. To further validate the effectiveness of DTNN and compare with state-of-the-art low energy implementation for edge vision, we design and implement DTNN based MLP image classifiers using off-the-shelf FPGAs. The results show that DTNN on the FPGA, with higher accuracy, could achieve orders of magnitude better energy consumption and latency compared with the state-of-the-art low energy approaches reported that use ASIC chips.