Explaining Black-Box Machine Learning Predictions

Abstract:
Machine learning is at the forefront of many recent advances in science and technology, enabled in part by the sophisticated models and algorithms that have been recently introduced. However, as a consequence of this complexity, machine learning essentially acts as a black-box as far as users are concerned, making it incredibly difficult to understand, predict, or "trust" their behavior.

In this talk, I will describe approaches to explain the predictions of ANY classifier in an interpretable and faithful manner. I will present examples of explanations for a variety of data types, such as images, text, and tabular data, and complex classifiers, such as random forests and deep neural networks. I will also present quantitative evaluation with human subjects on scenarios that require trust: deciding if one should trust a prediction, choosing between two algorithms, improving an untrustworthy algorithm, and predicting the behavior of the classifier.

Agenda:
6:15 - 6:45
Food, drinks and networking
6:45 - 7:00
Announcements on DataScience platform - CSO William Merchan
7:00 - 8:00
Professor Sameer's talk on Explaining Black-Box Machine Learning Predictions
8:00 - 8:30 Striking a Balance Between Model Performance and Interpretability using Skater - Pramit Choudhary( Lead DataScientist )
8:30 - 8:45 Raffle - Win a ticket to PyData, Seattle
8:45 - 9:15 More networking for interested folks

Bio:

https://secure.meetupstatic.com/photos/event/4/1/f/6/600_461656886.jpeg

Dr. Sameer Singh is an Assistant Professor of Computer Science at the University of California, Irvine. He is working on large-scale and interpretable machine learning applied to information extraction and natural language processing. Sameer was a Postdoctoral Research Associate at the University of Washington. He received his PhD from the University of Massachusetts, Amherst in 2014, during which he also worked at Microsoft Research, Google Research, and Yahoo! Labs on massive-scale machine learning. He was awarded the Adobe Research Data Science Faculty Award, was selected as a DARPA Riser, won the grand prize in the Yelp dataset challenge, and received the Yahoo! Key Scientific Challenges fellowship. Sameer has published at top-tier machine learning and natural language processing conferences and workshops. (http://sameersingh.org (http://sameersingh.org/))

Parking & Other Info:

• Enter from Hannum Ave into the parking structure for the 200-300 buildings
• Once you park, exit the parking structure on P4 near the elevators.
• Head across the courtyard veering slightly to the right and you will see the 200 building. If you find the 300 or 100 buildings you are in the wrong place. Take elevator to the second floor.
• Parking is free in the lot, after 6:30 PM the gates are open so people are free to leave.