Interactive Task Learning & Exploring and Understanding Large, Static and Dynamic Graphs

John Laird: The goal of my research is to develop a general cognitive architecture that supports human-level behavior. One capability of humans that is missing in artificial agents is interactive task learning, where an agent learns new tasks from natural interactions with a human instructor. This contrasts with approaches where an agent already has some internal representation of the task, but learns how to do the task well, as well as approaches that rely on other modalities such as demonstration that restrict the complexity of task that can be learned. Interactive task learning requires the integration of many areas of AI including natural language processing, dialog management, object recognition and perception, actuation, language and concept grounding, spatial reasoning, knowledge representation, and general problem solving. Our approach builds on prior research on cognitive architecture (Soar) that provides the necessary representation, processing, and learning mechanisms. Our approach emphasizes mixed initiative interaction, where the human provides advice and information, and the agent actively asks questions to acquire the knowledge it needs. Moreover, the agent learns by being situated in the task with the instructor, and it attempts to perform the task as it gains knowledge.

Danai Koutra: Networks naturally capture a host of real-world interactions, spanning from friendships to brain activity. But, given a massive graph, such as the Facebook social network, what can be learned about its structure? Where should people's attention be directed? The focus of my group is on scalable and principled methods that empower the end users to understand their interconnected data when they want to know broadly "what's in the data". The problems we tackle include summarization, similarity, alignment, and anomaly detection in graphs with several millions of nodes and edges. In this talk I will present our ongoing work on: (1) creating graph summaries which disentangle the complex connectivity patterns in static and dynamic graphs, and give interesting discoveries in real-world graphs; and (2) a scalable, interactive graph analytics platform, which aids the user in making sense of her data by summarizing important graph properties and directing attention to "anomalous' behaviors.
John Laird: John E. Laird is the John L. Tishman Professor of Engineering at the University of Michigan, where he has been since 1986. He received his Ph.D. in Computer Science from Carnegie Mellon University in 1983 working with Allen Newell. From 1984 to 1986, he was a member of research staff at . He is one of the original developers of the Soar architecture and leads its continued evolution. He was a founder of Soar Technology, Inc. and he is a Fellow of AAAI, AAAS, ACM, and the Cognitive Science Society.

Danai Koutra: Danai Koutra is an Assistant Professor in Computer Science and Engineering at the University of Michigan, Ann Arbor. Her group's research focuses on scalable methods that help the end users to understand their networked data and find interesting patterns. Broadly, her interests include large-scale graph mining, graph summarization, graph similarity and matching, and anomaly detection. Danai's research has been applied to social, collaboration and web networks, as well as brain connectivity graphs. She has numerous papers in top data mining conferences, including 2 award-winning papers, and 7 current and pending patents on bipartite graph alignment. Her work has also been covered by the popular press, such as the MIT Technology Review. Danai has worked at IBM Research, Microsoft Research, and Technicolor Research Labs. She earned her Ph.D. and M.S. in Computer Science from Carnegie Mellon University in 2015 and her diploma in Electrical and Computer Engineering at the National Technical University of Athens in 2010.