Stay Informed:
Baskin Engineering COVID-19 Information and Resources
Campus Roadmap to Recovery
Zoom Links: Zoom Help | Teaching with Zoom | Zoom Quick Guide

Reliable Decision Making in the Open World

Speaker Name: 
Sandhya Saisubramanian
Speaker Title: 
PhD Candidate
Speaker Organization: 
University of Massachusetts Amherst
Start Time: 
Friday, March 12, 2021 - 11:00am
End Time: 
Friday, March 12, 2021 - 12:15pm
Location: 
Via Zoom Link
Organizer: 
Lise Getoor

Abstract:

 

The increased deployment of artificial intelligence (AI) systems and their broad societal impacts brings along new challenges, particularly ensuring that these systems operate reliably in the open world. Due to the practical challenges in data collection and precise model specification, AI systems often operate based on incomplete information, which may lead to undesirable consequences. In this talk, I will present three lines of work that enable reliable decision making under uncertainty, resource constraints, and incomplete information: (i) mitigating the negative side effects of agent actions, while providing bounded-performance guarantees; (ii) decision making with imperfect information about goal states; and (iii) large-scale decision making in urban settings.

 

Bio:

 

Sandhya Saisubramanian is a Ph.D. candidate in Computer Science at the University of Massachusetts Amherst, advised by Shlomo Zilberstein. She is expected to graduate in June 2021. Her research focuses  on developing general techniques for reliable decision making in autonomous systems that operate in the open world. Her recent work on negative side effects was recognized with a Distinguished Paper Award at IJCAI 2020. She was also selected as a Rising Star in EECS by UC Berkeley in 2020.

As a gentle reminder, please respect the privacy of faculty recruitment by not sharing the candidate status of our guests with others outside of our organization.

 Zoom Link:   https://ucsc.zoom.us/j/97097761781?pwd=akhTVzJOWU9WQmZpTVp3RWszR012QT09

Event Type: 
Event