At this week's proseminar, we are hosting Patrick Hall. Patrick has done extensive work on explainable and responsible machine learning, a highly popular topic both in academia and industry.
This is a great learning opportunity for everyone, so the session is open to the whole StaSci department.
Increasing Trust and Interpretability in Machine Learning with Model Debugging
Prediction by machine learning models is fundamentally the execution of a computer program. In this case, the rules of the computer program are learned by the computer itself from training data instead of being programmed by a human. Like all good programs, machine learning models should be debugged to discover and remediate errors. When the debugging process increases accuracy in holdout data, increases transparency into model mechanisms, decreases or identifies hackable attack surfaces, or decreases disparate impact this debugging process also enhances trust and interpretability in model mechanisms and predictions. This text discusses several standard techniques in the context of model debugging: disparate impact, residual, and sensitivity analysis and introduces novel applications such as global and local explanation of model residuals.
Patrick Hall is principal scientist at bnh.ai, a D.C.-based law firm focused on AI and data analytics. He also serves as a visiting assistant professor of decision sciences at the George Washington University School of Business (GWSB), where his teaching and research focus on data mining, machine learning, and the responsible use of these technologies. Among other academic and technology media writing, Patrick is the primary author of popular e-books on explainable and responsible machine learning.
Before co-founding bnh.ai, Patrick led responsible AI efforts at H2O.ai, a leading machine learning software firm. His work at H2O.ai resulted in one of the world's first commercial solutions for explainable and fair machine learning. Prior to joining H2O.ai, Patrick held global customer-facing and R&D roles at SAS, where he authored multiple patents in automated market segmentation using novel clustering methods and deep learning. While at SAS, he also became the 11th person worldwide to become a Cloudera certified data scientist.
Patrick studied computational chemistry at the University of Illinois before graduating from the Institute for Advanced Analytics at North Carolina State University.