arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.09453cs.LGcs.ARcs.CL
收藏

权重块稀疏性:训练、编译与AI引擎加速器

Weight Block Sparsity: Training, Compilation, and AI Engine Accelerators

Xilinx/AMD(赛灵思/超威半导体)

Paolo D'Alberto, Taehee Jeong, Akshai Jain, Shreyas Manjunath, Mrinal Sarmah, Samuel Hsu, Yaswanth Raparti, Nitesh Pipralia

AI总结:

该研究提出了一套支持8x8权重块稀疏性的垂直系统,涵盖训练、编译与硬件加速,在Resnet50等模型上实现了精度损失极小情况下的推理加速,验证了软硬件协同的必要性。

摘要:

当前,规模越来越大的深度神经网络(DNN)正被不断开发、训练和投入使用。这类网络需要大量计算资源,给高端设备和算力有限的设备都带来了负担。我们的解决方案是实现权重块稀疏性——一种对硬件友好的结构化稀疏性。通过将预训练DNN模型的卷积层和全连接层参数的特定部分置零,我们能够高效加速DNN的推理过程,进而带来内存占用更小、通信速度更快、运算量更少的效果。\n 我们的工作提出了一套垂直系统,该系统支持在单块GPU上在合理时间内训练卷积权重与矩阵乘法权重,以利用8x8块稀疏性。编译器能够识别这种稀疏性,既将其用于数据压缩,也用于将计算拆分到不同线程中。这类块能够充分利用空间局部性与时间局部性,为快速向量运算和内存复用铺平了道路。将该系统应用于Resnet50模型时,我们在精度损失极小的情况下将权重规模缩减了一半,实现了2倍的推理速度提升。我们还将展示针对AIE2配置集(AMD Versal FPGA),基于Resnet50、Inception V3和VGG16通过精准完整的代码生成得到的性能预估,以此证明硬件覆盖设计与软件栈在机器学习应用的编译和执行中所需的协同作用。

英文摘要:

Nowadays, increasingly larger Deep Neural Networks (DNNs) are being developed, trained, and utilized. These networks require significant computational resources, putting a strain on both advanced and limited devices. Our solution is to implement {\em weight block sparsity}, which is a structured sparsity that is friendly to hardware. By zeroing certain sections of the convolution and fully connected layers parameters of pre-trained DNN models, we can efficiently speed up the DNN's inference process. This results in a smaller memory footprint, faster communication, and fewer operations. Our work presents a vertical system that allows for the training of convolution and matrix multiplication weights to exploit 8x8 block sparsity on a single GPU within a reasonable amount of time. Compilers recognize this sparsity and use it for both data compaction and computation splitting into threads. Blocks like these take full advantage of both spatial and temporal locality, paving the way for fast vector operations and memory reuse. By using this system on a Resnet50 model, we were able to reduce the weight by half with minimal accuracy loss, resulting in a two-times faster inference speed. We will present performance estimates using accurate and complete code generation for AIE2 configuration sets (AMD Versal FPGAs) with Resnet50, Inception V3, and VGG16 to demonstrate the necessary synergy between hardware overlay designs and software stacks for compiling and executing machine learning applications.

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。

↑