arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.02691cs.DCcs.LG
收藏

并行和分布式大模型深度学习训练系统

Systems for Parallel and Distributed Large-Model Deep Learning Training

University of California, San Diego(加州大学圣迭戈分校)

Kabir Nagrecha

摘要:

深度学习(DL)已经改变了各种领域的应用,包括计算机视觉、自然语言处理和表格数据分析。对改进的DL模型准确性的研究已经导致从业者探索越来越大的神经架构,最近的一些Transformer模型跨越了数千亿个可学习参数。这些设计为DL空间引入了新的规模驱动系统挑战,例如内存瓶颈,运行时效率低下以及模型开发的高成本。解决这些问题的努力已经探索了诸如神经架构的并行化、跨存储层次结构溢出数据以及存储器高效数据表示之类的技术。本次调查将探讨大型培训系统的前景,重点介绍主要挑战以及用于解决这些挑战的各种技术。

英文摘要:

Deep learning (DL) has transformed applications in a variety of domains, including computer vision, natural language processing, and tabular data analysis. The search for improved DL model accuracy has led practitioners to explore increasingly large neural architectures, with some recent Transformer models spanning hundreds of billions of learnable parameters. These designs have introduced new scale-driven systems challenges for the DL space, such as memory bottlenecks, poor runtime efficiency, and high costs of model development. Efforts to address these issues have explored techniques such as parallelization of neural architectures, spilling data across the memory hierarchy, and memory-efficient data representations. This survey will explore the large-model training systems landscape, highlighting key challenges and the various techniques that have been used to address them.

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。

↑