GOBO: Quantizing Attention-Based NLP Models for Low Latency and Energy Efficient Inference

Attention-based models have demonstrated remarkable success in various natural language understanding tasks. However, efficient execution remains a challenge for these models which are memory-bound due to their massive number of parameters. We present GOBO, a model quantization technique that compresses the vast majority (typically 99.9%) of the 32-bit floating-point parameters of state-of-the-art BERT models and their variants to 3 bits while maintaining their accuracy. Unlike other quantization methods, GOBO does not require fine-tuning nor retraining to compensate for the quantization error. We present two practical hardware applications of GOBO. In the first GOBO reduces memory storage and traffic and as a result inference latency and energy consumption. This GOBO memory compression mechanism is plug-in compatible with many architectures; we demonstrate it with the TPU, Eyeriss, and an architecture using Tensor Cores-like units. Second, we present a co-designed hardware architecture that also reduces computation. Uniquely, the GOBO architecture maintains most of the weights in 3b even during computation, a property that: (i) makes the processing elements area efficient, allowing us to pack more compute power per unit area, (ii) replaces most multiply-accumulations with additions, and (iii) reduces the off-chip traffic by amplifying on-chip memory capacity.

I. Introduction

Modern computing platforms, from servers to mobile and embedded devices, are energy constrained. Techniques for improving their energy efficiency can yield several benefits. They can reduce the energy footprint of data centers and as a result operating costs and environmental impact. They can increase uptime for mobile devices, and they can boost the capability of all systems by allowing them to perform more computations per unit of time. Improving energy efficiency is imperative for deep learning workloads as they are particularly compute and memory intensive. More so, given that the trend has been towards neural network models that require more computations and memory.