SHF: Conference: Hardware and Algorithms for Learning On-a-chip; November 5, 2015; Austin, TX

SHF：会议：片上学习的硬件和算法；

• 批准号: 1545974
• 负责人: Yu Cao
• 金额: $ 1.5万
• 依托单位: Arizona State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-07-15 至 2016-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1545974&HistoricalAwards=false
• 关键词: SHF; Conference; Hardware; Algorithms; Learning

Support for student participation in a workshop to establish a forum for discussion of the current practices, as well as future research needs in three interrelated subareas of brain related sciences is provided through this grant. The three specific subareas that the workshop will discuss are: hardware acceleration of on-chip learning; machine learning algorithms and neuro-inspired information processing; and nano-electronic design for learning, with the focus on performance and energy efficiency. The workshop will consist of three sessions and will be followed by a panel discussion, and will end with a poster session to present student work. The conference will not only provide a venue for researchers to address problems but also a platform for graduate students and others to learn about cutting-edge research by attending invited talks, presentations, tutorials in the general area.The outcome of this workshop include a webpage hosted by the Arizona State University to disseminate the program and discussions, as well as a report that clearly identifies major research challenges and needs in computational neurosciences, machine learning, and hardware design. The papers and posters will be published through the workshop proceedings. These results will illustrate new and exciting inter-disciplinary research problems of interest to the related communities, and articulate a vision of conducting further research in this area. The final workshop report will also be submitted to NSF as an annual report and will be released online for sharing with the research community in general.

通过这笔赠款，支持学生参加研讨会，建立一个论坛，讨论当前的实践以及脑相关科学三个相互关联的子领域的未来研究需求。研讨会将讨论的三个具体子领域是：片上学习的硬件加速；机器学习算法和神经启发信息处理；以及纳米电子设计学习，重点关注性能和能源效率。研讨会将由三场会议组成，随后将进行小组讨论，并以展示学生作品的海报会议结束。会议不仅为研究人员提供解决问题的场所，而且为研究生和其他人提供一个平台，通过参加一般领域的受邀演讲、演讲、教程来了解前沿研究。本次研讨会的成果包括一个网页由亚利桑那州立大学主办，旨在传播该计划和讨论，以及一份报告，该报告明确指出了计算神经科学、机器学习和硬件设计方面的主要研究挑战和需求。论文和海报将通过研讨会论文集发布。这些结果将说明相关社区感兴趣的新的、令人兴奋的跨学科研究问题，并阐明在该领域进行进一步研究的愿景。最终的研讨会报告也将作为年度报告提交给 NSF，并将在线发布以与整个研究界共享。