• 学生活动

Fusion of Supervised Learning and Reinforcement Learning for Dynamic Treatment Recommendation

主讲人: 荆炳义
主讲人简介:

荆炳义,南方科技大学统计与数据科学系讲席教授。曾获国家自然科学奖二等奖,教育部高等学校自然科学奖二等奖。美国统计学会会士(2018)、数理统计学会会士 (2018)、国际统计学会当选会士(2006)。泛华统计协会理事会成员,中国现场统计学会多元分析委员会理事长,并先后分别担任七家国际学术期刊副主编。研究兴趣包括概率统计、计量经济、网络数据、强化学习、及生物信息等领域,有多项开创性研究和突破性科研成果。在Annals of Statistics、Annals of Probability、Biometrika、Journal of the American Statistical Association、Journal of Econometrics等顶级期刊发表论文100余篇,论文引超过3700余次。同时与产业界有着密切的交流与合作。

主持人: 洪永淼
简介:

Electronic health records (EHR) have provided a great opportunity to exploit personalized health data to optimize clinical decision making and achieve personalized treatment recommendation. In this talk, we explore how AI could help physicians in prescribing medicines for patients with multi-morbidity (i.e., co-occurrence of two or more diseases). Both Supervised Learning (SL) and Reinforcement Learning (RL) have been employed for this purpose, but with their own drawbacks. For instance, SL relies highly on the clinical guideline and doctors personal experience while RL may produce unacceptable medications due to lack of the supervision from doctors. In this talk, we propose a novel SAVER framework by fusing RL and SL, where RL learns the optimal policy and SL gives a regularization to avoid unacceptable risks. Our experiments show that our SAVER framework can provider more accuracy treatment recommendation than the existing methods.

时间: 2022-11-29 (Tuesday) 16:30-18:00
地点: 中科院数学与系统科学研究院南楼N204 (线下主会场)、厦大经济楼N302(线下分会场)、腾讯会议:37586125504
讲座语言: 中文
期数: “邹至庄讲座”杰出学者论坛(第13期)
主办单位: 中国科学院大学经济与管理学院、中国科学院预测科学研究中心、厦门大学邹至庄经济研究院、NSFC"计量建模与经济政策研究”基础科学中心
承办单位:
类型:
系列讲座
联系人信息: 许老师,电话:0592-2182991,邮箱:ysxu@xmu.edu.cn