MAx-DNN: Multi-Level Arithmetic Approximation for Energy-Efficient DNN Hardware Accelerators

TL;DR

This paper proposes a multi-level approximation method for DNN hardware accelerators that significantly reduces energy consumption while maintaining high accuracy, by strategically applying approximate multipliers across network layers.

Contribution

It introduces a systematic layer-, filter-, and kernel-level approach to applying approximate multipliers for energy-efficient DNN hardware, demonstrating substantial energy savings with minimal accuracy loss.

Findings

Up to 54% energy savings with 4% accuracy loss

Achieves 2x energy savings over state-of-the-art approximations

Effective application of approximate multipliers across network levels

Abstract

Nowadays, the rapid growth of Deep Neural Network (DNN) architectures has established them as the defacto approach for providing advanced Machine Learning tasks with excellent accuracy. Targeting low-power DNN computing, this paper examines the interplay of fine-grained error resilience of DNN workloads in collaboration with hardware approximation techniques, to achieve higher levels of energy efficiency. Utilizing the state-of-the-art ROUP approximate multipliers, we systematically explore their fine-grained distribution across the network according to our layer-, filter-, and kernel-level approaches, and examine their impact on accuracy and energy. We use the ResNet-8 model on the CIFAR-10 dataset to evaluate our approximations. The proposed solution delivers up to 54% energy gains in exchange for up to 4% accuracy loss, compared to the baseline quantized model, while it provides 2x energy gains with better accuracy versus the state-of-the-art DNN approximations.