Exploring Quantization and Mapping Synergy in Hardware-Aware Deep Neural Network Accelerators

TL;DR

This paper demonstrates that combining mixed quantization schemes with optimized mapping strategies in hardware-aware CNN accelerators can significantly improve energy efficiency and memory usage without sacrificing accuracy.

Contribution

It introduces a novel extension of the Timeloop tool for mixed quantization support and proposes an optimization algorithm for layer-specific quantization and mapping.

Findings

Energy savings up to 37% on MobileNet models

Supports mixed quantization in mapping tool

Improves accuracy-energy-memory trade-offs

Abstract

Energy efficiency and memory footprint of a convolutional neural network (CNN) implemented on a CNN inference accelerator depend on many factors, including a weight quantization strategy (i.e., data types and bit-widths) and mapping (i.e., placement and scheduling of DNN elementary operations on hardware units of the accelerator). We show that enabling rich mixed quantization schemes during the implementation can open a previously hidden space of mappings that utilize the hardware resources more effectively. CNNs utilizing quantized weights and activations and suitable mappings can significantly improve trade-offs among the accuracy, energy, and memory requirements compared to less carefully optimized CNN implementations. To find, analyze, and exploit these mappings, we: (i) extend a general-purpose state-of-the-art mapping tool (Timeloop) to support mixed quantization, which is not currently available; (ii) propose an efficient multi-objective optimization algorithm to find the most suitable bit-widths and mapping for each DNN layer executed on the accelerator; and (iii) conduct a detailed experimental evaluation to validate the proposed method. On two CNNs (MobileNetV1 and MobileNetV2) and two accelerators (Eyeriss and Simba) we show that for a given quality metric (such as the accuracy on ImageNet), energy savings are up to 37% without any accuracy drop.