Dustin Arthur Smith, MSc

Dustin Arthur Smith, MSc is a Ph.D. student in the Media Lab, working at the intersection of planning and natural language processing. His advisors are Henry Lieberman and Marvin Minsky. He is also a member of the MIT Mind Machine Project which aims to reconcile natural intelligence with machine intelligence, and in doing so develop and engineer a class of intelligent machines.
 
His research goal is to make computers understand English in a similar functional capacity as people. This ambition treads many academic topics: machine reading and story understanding, event structures and lexical semantics, semantic role labeling, statistical relational learning, sequence mining, event recognition and extraction, planning, plan recognition, metacognition, and self-modeling.
 
Dustin authored Notes on Problem Reformulation and Directions for Artificial Intelligence, and coauthored The Why UI: Using Goal Networks to Improve User Interfaces, Learning Hierarchical Plans by Reading Simple English Narratives, An Interface for Targeted Collection of Common Sense Knowledge Using a Mixture Model, Common Consensus: a web-based game for collecting commonsense goals, Action planning with commonsense knowledge, Recognizing and using goals in event management, and Unsupervised learning of common sense event structures from simple English stories.
 
Dustin earned his BSc in Computer Science (with a minor in Neuroscience) at Wake Forest University in 2005. He earned his MSc in Media Arts and Sciences from MIT in 2007 with the thesis EventMinder: A Personal Calendar Assistant That Understands Events.