Information Retrieval Seminar: Neural Retrieval Models and the Next Generation of Information Access

Abstract: Information access systems, including search engines, question answering, and recommender systems, have made significant impacts on people's daily lives. In this talk, I will highlight the current major challenges in developing the next generation of information access from both modeling and application perspectives. I will particularly focus on efficient and effective neural retrieval models to address these challenges. This includes methods for learning inverted indexes for efficient high-recall retrieval and contextual representations for efficient high-precision retrieval. Furthermore, I will briefly discuss how recent progress in neural retrieval models enables us to bridge the gap between search and recommendation systems. Finally, I will talk about conversational information access with a focus on mixed-initiative interactions. This includes asking clarifying questions in response to ambiguous or faceted queries. I will introduce methods for the generation and evaluation of clarification responses in conversational information access systems.

Bio: Hamed Zamani is a Researcher at Microsoft AI Research, working on a wide range of problems on neural information retrieval and conversational search. He obtained his Ph.D. in Computer Science from the University of Massachusetts Amherst under the direction of W. Bruce Croft. Hamed is an active member of the information retrieval (IR) community and is broadly interested in developing statistical and machine learning models with applications to information access systems, including search engines, question answering, and recommender systems. He was a co-chair for multiple workshops at SIGIR and RecSys, and has served as a Senior PC or PC member for SIGIR, WSDM, WWW, RecSys, CIKM, and ICTIR.

A reception for attendees will be held at 3:30 p.m. in CS 150.