Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning
About
We consider the problem of model compression for deep neural networks (DNNs) in the challenging one-shot/post-training setting, in which we are given an accurate trained model, and must compress it without any retraining, based only on a small amount of calibration input data. This problem has become popular in view of the emerging software and hardware support for executing models compressed via pruning and/or quantization with speedup, and well-performing solutions have been proposed independently for both compression approaches. In this paper, we introduce a new compression framework which covers both weight pruning and quantization in a unified setting, is time- and space-efficient, and considerably improves upon the practical performance of existing post-training methods. At the technical level, our approach is based on an exact and efficient realization of the classical Optimal Brain Surgeon (OBS) framework of [LeCun, Denker, and Solla, 1990] extended to also cover weight quantization at the scale of modern DNNs. From the practical perspective, our experimental results show that it can improve significantly upon the compression-accuracy trade-offs of existing post-training methods, and that it can enable the accurate compound application of both pruning and quantization in a post-training setting.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Image Classification | ImageNet-1k (val) | Top-1 Accuracy75.72 | 512 | |
| Image Classification | ImageNet | Top-1 Accuracy75.64 | 324 | |
| Image Classification | ImageNet (val) | Accuracy75.2 | 300 | |
| Object Detection | COCO | AP50 (Box)66.14 | 190 | |
| Image Classification | ImageNet (val) | Top-1 Accuracy75.5 | 118 | |
| Question Answering | SQuAD v1.1 | F187.81 | 79 | |
| Question Answering | SQuAD v1.1 (val) | F1 Score86.97 | 70 | |
| Image Classification | ImageNet-1k (val) | Accuracy71.47 | 5 |