ISSIP

ISSIP Cognitive Systems Institute Group Speaker Series : April 25, 10:30am US Eastern Time

“Event memory for explainable autonomy”
Dongkyu Choi, A*STAR & David Ménager, University of Kansas

When:  Thursday, April 25, 10:30 am US Eastern.

Zoom Detail Below

Background:

Dongkyu Choi is a senior scientist at Singapore’s Agency for Science, Technology and Research. His current research focuses on artificial agents that collaborate with human counterparts on physical tasks. Based on his expertise in cognitive architectures, Dongkyu is leading the integration effort that brings together perception, action, dialogue, learning, and commonsense reasoning within a unified framework. He is experienced in designing, managing, and performing funded projects from various government agencies, including Defense Advanced Research Projects Agency, Office of Naval Research, Naval Research Laboratory, and Korea Institute of Science and Technology. Previously, Dongkyu was a member of faculty at the University of Kansas and a postdoctoral researcher at the University of Illinois at Chicago. He received his MS and PhD degrees from Stanford University and his BS degree from Seoul National University.

 

David Menager is a PhD student at the University of Kansas. His research uses computational models of event memory as a core technology for answering research questions about artificial intelligence. His current interest is to understand the role of event memory for explainable AI in single-agent and multi-agent domains. David spent last summer as an intern at Lockheed Martin’s Skunk Works researching machine learning. In his free time, David likes to produce music and go running.

Many of intelligent agents available today are not able to remember their past interactions with users and cannot explain their decision making processes. As a remedy to these issues, we propose a novel theory of event memory that can provide an infrastructure for such capacities. The psychologically plausible theory enables the agent to store both individual instances and probabilistic schemas of personal events. Using the contents of this memory, intelligent agents can interact with their users in a contextually relevant manner and provide explanations and justifications for their decisions. In this talk, they describe representation and processes of this framework and discuss its implications to intelligent agency.

Zoom meeting Link: https://zoom.us/j/7371462221

Zoom Callin: (415) 762-9988 or (646) 568-7788 Meeting id 7371462221

Zoom International Numbers: https://zoom.us/zoomconference

Check the website in case the date or time changes: http://cognitive-science.info/community/weekly-update/  & where you can find slides & recordings of prior calls)

Please retweet https://twitter.com/sumalaika/status/1120268098228183041

Join LinkedIn Group https://www.linkedin.com/groups/6729452