arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.04092cs.LG
收藏

通过双层优化推进模型剪枝

Advancing Model Pruning via Bi-level Optimization

Michigan State University(密歇根州立大学);IBM Research(IBM研究院);Northeastern University(东北大学);University of Texas at Austin(德克萨斯大学奥斯汀分校);University of Minnesota, Twin Cities(明尼苏达大学双城分校)

Yihua Zhang, Yuguang Yao, Parikshit Ram, Pu Zhao, Tianlong Chen, Mingyi Hong, Yanzhi Wang, Sijia Liu

AI总结:

本文将模型剪枝表述为具有双线性结构的双层优化问题,提出 BiP,以一阶优化效率取得优于或持平 IMP 的中奖彩票质量,并在多种架构和数据集上实现 2–7 倍加速。

摘要:

实际应用中的部署约束使得大规模深度学习模型的剪枝成为必要,即提升其权重稀疏性。正如彩票假设(Lottery Ticket Hypothesis,LTH)所阐明的,剪枝也具有提升模型泛化能力的潜力。在 LTH 的核心中,迭代幅度剪枝(iterative magnitude pruning,IMP)是成功找到“中奖彩票”的主流剪枝方法。然而,随着目标剪枝率提高,IMP 的计算成本会增长到令人难以承受。为降低计算开销,人们开发了多种高效的“一次性”剪枝方法,但这些方案通常无法找到像 IMP 那样好的中奖彩票。这就提出了一个问题:如何弥合剪枝精度与剪枝效率之间的差距?为解决该问题,我们推进模型剪枝的算法研究。具体而言,我们从一个全新的新颖视角——双层优化(bi-level optimization,BLO)——来表述剪枝问题。我们表明,BLO 解释为高效实现 IMP 中使用的剪枝—再训练学习范式提供了具有技术依据的优化基础。我们还表明,所提出的面向双层优化的剪枝方法(称为 BiP)是一类具有双线性问题结构的特殊 BLO 问题。通过利用这种双线性,我们从理论上证明 BiP 可以像一阶优化一样容易求解,从而继承计算效率。通过在 5 种模型架构和 4 个数据集上对结构化与非结构化剪枝进行大量实验,我们证明 BiP 在大多数情况下能够找到比 IMP 更好的中奖彩票,并且在计算上与一次性剪枝方案同样高效;在相同模型精度和稀疏性水平下,相比 IMP 实现 2–7 倍加速。

英文摘要:

The deployment constraints in practical applications necessitate the pruning of large-scale deep learning models, i.e., promoting their weight sparsity. As illustrated by the Lottery Ticket Hypothesis (LTH), pruning also has the potential of improving their generalization ability. At the core of LTH, iterative magnitude pruning (IMP) is the predominant pruning method to successfully find 'winning tickets'. Yet, the computation cost of IMP grows prohibitively as the targeted pruning ratio increases. To reduce the computation overhead, various efficient 'one-shot' pruning methods have been developed, but these schemes are usually unable to find winning tickets as good as IMP. This raises the question of how to close the gap between pruning accuracy and pruning efficiency? To tackle it, we pursue the algorithmic advancement of model pruning. Specifically, we formulate the pruning problem from a fresh and novel viewpoint, bi-level optimization (BLO). We show that the BLO interpretation provides a technically-grounded optimization base for an efficient implementation of the pruning-retraining learning paradigm used in IMP. We also show that the proposed bi-level optimization-oriented pruning method (termed BiP) is a special class of BLO problems with a bi-linear problem structure. By leveraging such bi-linearity, we theoretically show that BiP can be solved as easily as first-order optimization, thus inheriting the computation efficiency. Through extensive experiments on both structured and unstructured pruning with 5 model architectures and 4 data sets, we demonstrate that BiP can find better winning tickets than IMP in most cases, and is computationally as efficient as the one-shot pruning schemes, demonstrating 2-7 times speedup over IMP for the same level of model accuracy and sparsity.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑