Hardware-Centric AutoML for Mixed-Precision Quantization

by   Kuan Wang, et al.

Model quantization is a widely used technique to compress and accelerate deep neural network (DNN) inference. Emergent DNN hardware accelerators begin to support mixed precision (1-8 bits) to further improve the computation efficiency, which raises a great challenge to find the optimal bitwidth for each layer: it requires domain experts to explore the vast design space trading off among accuracy, latency, energy, and model size, which is both time-consuming and sub-optimal. Conventional quantization algorithm ignores the different hardware architectures and quantizes all the layers in a uniform way. In this paper, we introduce the Hardware-Aware Automated Quantization (HAQ) framework which leverages the reinforcement learning to automatically determine the quantization policy, and we take the hardware accelerator's feedback in the design loop. Rather than relying on proxy signals such as FLOPs and model size, we employ a hardware simulator to generate direct feedback signals (latency and energy) to the RL agent. Compared with conventional methods, our framework is fully automated and can specialize the quantization policy for different neural network architectures and hardware architectures. Our framework effectively reduced the latency by 1.4-1.95x and the energy consumption by 1.9x with negligible loss of accuracy compared with the fixed bitwidth (8 bits) quantization. Our framework reveals that the optimal policies on different hardware architectures (i.e., edge and cloud architectures) under different resource constraints (i.e., latency, energy, and model size) are drastically different. We interpreted the implication of different quantization policies, which offer insights for both neural network architecture design and hardware architecture design.


HAQ: Hardware-Aware Automated Quantization

Model quantization is a widely used technique to compress and accelerate...

Design Automation for Efficient Deep Learning Computing

Efficient deep learning computing requires algorithm and hardware co-des...

Generative Design of Hardware-aware DNNs

To efficiently run DNNs on the edge/cloud, many new DNN inference accele...

Survey and Benchmarking of Precision-Scalable MAC Arrays for Embedded DNN Processing

Reduced-precision and variable-precision multiply-accumulate (MAC) opera...

A mixed signal architecture for convolutional neural networks

Deep neural network (DNN) accelerators with improved energy and delay ar...

Automatic low-bit hybrid quantization of neural networks through meta learning

Model quantization is a widely used technique to compress and accelerate...

APQ: Joint Search for Network Architecture, Pruning and Quantization Policy

We present APQ for efficient deep learning inference on resource-constra...

Please sign up or login with your details

Forgot password? Click here to reset