arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.12922cs.AR
收藏

Griffin:重新思考深度学习架构的稀疏优化

Griffin: Rethinking Sparse Optimization for Deep Learning Architectures

Jong Hoon Shin, Ali Shafiee, Ardavan Pedram, Hamzah Abdel-Aziz, Ling Li, Joseph Hassoun

摘要:

本文研究了DNN加速器的设计空间权衡,旨在实现密集或稀疏激活/权重张量的所有四种组合的有竞争力的性能和效率指标。要做到这一点,我们系统地研究了支持稀疏的优化密集的核心顶部的开销。这些开销是基于指示乘法器如何从相邻乘法器或未来周期借用非零操作的参数来建模的。作为这种探索的结果,我们确定了一些有前途的设计,比以前的工作表现更好。我们的研究结果表明,即使是针对双稀疏性的最佳设计,当在单个稀疏模型上执行时,也会产生20%-30%的功率效率下降,即,仅具有稀疏权重或稀疏激活张量的那些。我们发现,当运行单个稀疏或密集模型时,可以重用同一核心的资源以保持高性能和效率。我们称之为混合架构格里芬。对于密集、仅加权稀疏、仅激活稀疏和双稀疏模型,Griffin的能效分别比最先进的稀疏架构高1.2、3.0、3.1和1.4倍。

英文摘要:

This paper examines the design space trade-offs of DNNs accelerators aiming to achieve competitive performance and efficiency metrics for all four combinations of dense or sparse activation/weight tensors. To do so, we systematically examine the overheads of supporting sparsity on top of an optimized dense core. These overheads are modeled based on parameters that indicate how a multiplier can borrow a nonzero operation from the neighboring multipliers or future cycles. As a result of this exploration, we identify a few promising designs that perform better than prior work. Our findings suggest that even the best design targeting dual sparsity yields a 20%-30% drop in power efficiency when performing on single sparse models, i.e., those with only sparse weight or sparse activation tensors. We found that one can reuse resources of the same core to maintain high performance and efficiency when running single sparsity or dense models. We call this hybrid architecture Griffin. Griffin is 1.2, 3.0, 3.1, and 1.4X more power-efficient than state-of-the-art sparse architectures, for dense, weight-only sparse, activation-only sparse, and dual sparse models, respectively.

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。

↑