Event Details

There has been a great deal of interest in Explainable Artificial Intelligence (XAI) recently. However, the research topic is not particularly new.   This talk will review the speaker’s mid-90s NSF funded work on Comprehensible Knowledge Discovery in Databases that was motivated in part by a collaboration with a neurologist on early diagnosis of dementia.  This research focused on biasing learning algorithms to produce models that are acceptable to practitioners.

The success of neural networks and deep learning in numerous problems including image classification has renewed interest in XAI.  Considerable progress has been made on exposing the inner workings of deep learning algorithms.  However, for the most part, the explanations produced by the current generation of XAI systems are unlike those produced by human experts.   The talk concludes with an analysis of expert’s classifying images and a discussion of recent results in collaboration with a radiologist at UCSD on differential diagnosis by analyzing chest X-rays of COVID-19 patients.

Speaker Bio

Michael Pazzani is a Distinguished Scientist at the Halıcıoğlu Data Science Institute at the University of California, San Diego. He conducts research in machine learning, explainable artificial intelligence, personalization, internet search, and recommendation systems.

Prior to UCSD, Dr. Pazzani was the Vice Chancellor for Research and Economic Development at the University of California, Riverside where he was also a professor of computer science with additional appointments in statistics and psychology.  He served as Vice President for Research at Rutgers University and the Director of the Information and Intelligence Systems Division at NSF.  He currently is a member of the Defense Science Board.

Dr. Pazzani received his Ph.D. in Computer Science from UCLA and was an assistant, associate, and full professor at the University of California, Irvine, where he also served as Chair of Information and Computer Science.