Open Compass: Accelerating the Adoption of AI in Open Research
Open Compass: Accelerating the Adoption of AI in Open Research
复制标题
DOI：
10.1145/3332186.3332253
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Paola A. Buitrago;Nicholas A. Nystrom
中科院分区：
文献类型：
--
作者：
Paola A. Buitrago;Nicholas A. Nystrom
人工智能（AI）在研究和工业领域具有巨大的潜力。人工智能应用比比皆是，并且正在迅速扩展，但人工智能的方法、性能和理解仍处于起步阶段。研究人员面临着一些棘手的问题，例如如何提高性能、可转移性、可靠性、可理解性，以及如何仅用有限的数据更好地训练人工智能模型。未来的进步取决于硬件加速器、软件框架、系统和架构的进步，以及在科学和人工智能领域之间创建跨领域专业知识。 Open Compass 是一个探索性研究项目，旨在对人工智能的先进工程测试平台 Compass Lab 进行学术试点研究，最终开发和发布最佳实践，造福广大科学界。 Open Compass 包括开发一个本体来描述现有和新兴人工智能硬件技术的复杂范围，以及识别代表训练深度学习模型的不同挑战的基准问题。然后，使用这些基准测试在替代的高级硬件解决方案架构中执行实验。在这里，我们介绍了 Open Compass 的方法，以及分析不同 GPU 类型、内存和拓扑对适用于图像处理的流行深度学习模型的影响的一些初步结果。
Artificial intelligence (AI) has immense potential spanning research and industry. AI applications abound and are expanding rapidly, yet the methods, performance, and understanding of AI are in their infancy. Researchers face vexing issues such as how to improve performance, transferability, reliability, comprehensibility, and how better to train AI models with only limited data. Future progress depends on advances in hardware accelerators, software frameworks, system and architectures, and creating cross-cutting expertise between scientific and AI domains. Open Compass is an exploratory research project to conduct academic pilot studies on an advanced engineering testbed for artificial intelligence, the Compass Lab, culminating in the development and publication of best practices for the benefit of the broad scientific community. Open Compass includes the development of an ontology to describe the complex range of existing and emerging AI hardware technologies and the identification of benchmark problems that represent different challenges in training deep learning models. These benchmarks are then used to execute experiments in alternative advanced hardware solution architectures. Here we present the methodology of Open Compass and some preliminary results on analyzing the effects of different GPU types, memory, and topologies for popular deep learning models applicable to image processing.