FAMES: Fast Approximate Multiplier Substitution for Mixed-Precision Quantized DNNs--Down to 2 Bits!

TL;DR

FAMES introduces a rapid approximate multiplier substitution technique for ultra-low bitwidth quantized DNNs, significantly reducing energy consumption while maintaining accuracy, and outperforming prior methods in speed.

Contribution

This work presents the first systematic method for applying approximate multipliers to 2-bit quantized DNNs, achieving substantial energy savings and faster computation.

Findings

28.67% average energy reduction in mixed-precision models

Accuracy loss kept under 1%

Up to 300x faster than previous genetic algorithm methods

Abstract

A widely-used technique in designing energy-efficient deep neural network (DNN) accelerators is quantization. Recent progress in this direction has reduced the bitwidths used in DNN down to 2. Meanwhile, many prior works apply approximate multipliers (AppMuls) in designing DNN accelerators to lower their energy consumption. Unfortunately, these works still assume a bitwidth much larger than 2, which falls far behind the state-of-the-art in quantization area and even challenges the meaningfulness of applying AppMuls in DNN accelerators, since a high-bitwidth AppMul consumes much more energy than a low-bitwidth exact multiplier! Thus, an important problem to study is: Can approximate multipliers be effectively applied to quantized DNN models with very low bitwidths? In this work, we give an affirmative answer to this question and present a systematic solution that achieves the answer: FAMES, a fast approximate multiplier substitution method for mixed-precision DNNs. Our experiments demonstrate an average 28.67% energy reduction on state-of-the-art mixed-precision quantized models with bitwidths as low as 2 bits and accuracy losses kept under 1%. Additionally, our approach is up to 300x faster than previous genetic algorithm-based methods.