Abstract:
In many fields one needs to build predictive models for a set of related machine learning tasks. Traditionally these tasks are treated independently and the inference is done separately for each task, which ignores inherent connections among the tasks. Multi-task learning aims to improve the generalization performance by building models for all tasks simultaneously, leveraging inherent relatedness of these tasks. In this seminar, we introduce what is multi-task learning and show how it can be applied to improve the predictive modeling in various application areas such as biomedical informatics.

Biography:
Jiayu Zhou is an assistant professor at Department of Computer Science and Engineering, Michigan State University. Before joining MSU, Jiayu was a staff research scientist at Samsung Research America, leading the industrial research on recommender systems and deep learning algorithms. Jiayu received his Ph.D. degree in computer science at Arizona State University in 2014, under the supervision of Professor Jieping Ye. Jiayu has a broad research interest in large-scale machine learning and data mining, and biomedical informatics. He has served as Senior Program Committee of IJCAI 2015. He also served as program committee members in premier conferences such as NIPS, ICDM, SDM, WSDM, ACML and PAKDD. Jiayu currently serves as an Associate Editor of Neurocomputing. Most of Jiayu's research has been published in top machine learning and data mining venues including NIPS, SIGKDD, ICDM and SDM. One of his papers has been selected for the best student paper award in ICDM 2014.

Host: