光华讲坛——社会名流与企业家论坛第 5834 期
(线上讲座)
主题:FEAT: an Flexible, Efficient, and Accurate Test Strategy for COVID-19
主讲人:香港科技大学荆炳义教授
主持人:统计学院 林华珍教授
时间:2020年7月3日(周五)14:00-15:00
直播平台及会议ID:腾讯会议,304 518 283
主办单位:统计研究中心、数据科学与商业智能联合实验室和统计学院 科研处
主讲人简介:
荆炳义, 现为香港科技大学数学系教授、统计科学中心主任。研究领域包括统计机器学习,金融计量,生物信息,概率与统计等,共发表论文90余篇。荆炳义曾获国家自然科学二等奖, 教育部高等学校自然科学二等奖。他是美国统计协会(ASA)和数理统计协会(IMS)会士,国际统计学会(ISI)当选会士。目前担任以下六家国际期刊副主编,包括Journal of Business & Economic Statistics; Canadian Journal of Statistics; Statistics and Its Interface; Science in China; Journal of Data Analysis; Statistical Theory and Related Fields。
详情请见其个人主页: http://www.math.ust.hk/~majing/
内容提要:
Early detection of COVID-19 proves to be critical in mitigating the spread of the virus. There are various tests in common use, based on nucleic acid detection, antibody detection, and/or CT scanning. However, efficiency and accuracy have always been an issue: some are accurate but time-consuming while others are cheap but lacking accuracy. In this talk, we propose a Flexible, Efficient and Accurate Test (FEAT), which can dramatically improve both efficiency and accuracy for any existing test. The strategy is simple to understand and easy to implement in practice, and potentially useful in many scenarios.
COVID-19的早期检测对减缓病毒传播至关重要。常用的检测方法有核酸检测、抗体检测和/或CT扫描。然而,效率和准确性一直是一个问题: 有些准确性很高,但很耗时,而有些便宜但缺乏准确性。 在本次报告中,我们提出了一种灵活,高效和准确的测试(FEAT),它可以极大地提高任何现有测试的效率和准确性。该策略易于理解,易于在实践中实现,并且在许多场景中可能有用。