Prof. Xi Chen
Assistant professor, New York University
Talk:
Sequential Analysis for Crowd Labeling and Ranking
Abstract:
This talk proposes new sequential analysis models and methods motivated by crowdsourcing applications. We consider both crowdsourced binary labeling tasks and ranking tasks, where the requestor needs to decide which worker (or which pair of objects) for labeling and when to stop collecting labels to save for budget. For the binary labeling tasks, we propose an adaptive sequential probability ratio test (Ada-SPRT), which obtains the optimal experiment selection rule, the optimal stopping time,and the optimal decision rule under a single Bayesian decision framework. For crowd ranking, due to the complex structure, the optimal ranking policy is hard to compute. To address this challenge, we develop an asymptotically optimal ranking policy that achieves the minimal Bayes risk as the sample size goes to infinity. Although this work is motivated by crowdsourcing applications, the proposed methods can be applied a wide range of online decision problems, where there is a need for balancing the quality of the solution and the cost of collecting data. This is a joint work with Xiaoou Li, Yunxiao Chen, Jingchen Liu, and Zhiliang Ying.
Biography:
Xi Chen is an assistant professor at Department of Information, Operations, and Management Sciences at Stern School of Business at New York University. Before that, he was a Postdoc in the group of Prof. Michael Jordan at UC Berkeley. He obtained his Ph.D. from the Machine Learning Department at Carnegie Mellon University (CMU); and his Masters degree in Operations Research from the Tepper School of Business at CMU.He studies machine learning for crowdsourcing and high-dimensional statistics. He also studies operations research/management problems, such as the process flexibility, and data-driven revenue management. He received Simons-Berkeley Research Fellowship, Google Faculty Research Award, and was featured in 2017 Forbes list of "30 Under30 in Science".
