*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Towards natural human-AI interactions in vision and language
Date: Thursday, November 29 2018
Time: 10:00AM - 11:30AM (ET)
Location: CCB 360B
Arjun Chandrasekaran
Ph.D. Student in Computer Science
School of Interactive Computing
Georgia Institute of Technology
http://www.prism.gatech.edu/~arjun9/
Committee:
Dr. Devi Parikh (Advisor, School of Interactive Computing, Georgia Institute of Technology)
Dr. Dhruv Batra (School of Interactive Computing, Georgia Institute of Technology)
Dr. Sonia Chernova (School of Interactive Computing, Georgia Institute of Technology)
Dr. Mohit Bansal (Computer Science Dept., University of North Carolina at Chapel Hill)
Abstract:
Inter-human interaction is a rich form of communication. Human interactions typically leverage a good theory of mind, involve pragmatics, story-telling, humor, sarcasm, empathy, sympathy, etc. Current human-AI interactions, however, lack many of these features that characterize inter-human interactions. Towards the goal of developing AI that can interact with humans naturally (similar to other humans), in this dissertation, I take steps towards studying aspects of humor, story-telling, and theory of (AI's) mind.
Specifically, I
1. Build computational models for humor manifested in static images, and contextual, multi-modal humor.
2. Introduce a picture-sequencing task where a computational model learns the correct temporal order of events in a story.
3. Evaluate different factors that influence the extent to which a lay person can predict the behavior of an AI, i.e., a person's theory of the AI's mind.
In proposed work, I will evaluate the extent to which interpretable AI approaches improve the overall performance of a human-AI team in a goal-driven, cooperative task.
