Learning to rank is a statistical learning task. The goal of it is to automatically construct a ranking model (function) using training data, such that the model can sort objects according to their degrees of relevance, preference, or importance defined in a specific application. Learning to rank has been receiving keen and growing interest in machine learning, data mining, information retrieval, and other fields in recent years, because of its importance, novelty, and far-reaching implication. In this talk, I will introduce our recent work on learning to rank, specifically the listwise approach to learning to rank. First, I will give a survey on the topic, and introduce the state-of-the-art pairwise approach. Next, I will point out the necessity of adopting a listwise approach, and introduce our proposed methods of ListNet and AdaRank, belonging to the category. The former is an algorithm for learning a probabilistic model. The latter is a Boosting algorithm for directly optimizing a listwise loss function.
Bio: Hang Li is a senior researcher and research manager at Microsoft Research Asia. His research areas include natural language processing, information retrieval, statistical machine learning, and data mining. He graduated from Kyoto University and earned his PhD from the University of Tokyo. http://research.microsoft.com/users/hangli/