This is a past event
At my dissertation defense, I presented Planning and Recognition Together Close the Interaction Loop (PReTCIL) as one of the first ways to integrate algorithms from artificial intelligence (AI) planning and plan, activity, and intent recognition into a single method for adaptive closed-loop interaction in autonomous systems. Despite PReTCIL working out mathematically and in computational evaluation, things went wrong when people interacted with an implementation. Making AI human-aware goes beyond theory and code because people are not simply data, easy to simulate via algorithms, or assumptions we can overlook---they are autonomous agents with their own goals, beliefs, preferences, and experiences. This talk introduces human-aware design for human-aware AI as a means to bring human factors and human-computer/robot interaction principles into the development of AI algorithms that interact with people. Those involved in the research, implementation, and deployment of such machines' AI systems need to understand people and consider them as much as the systems will. Through the lens of human-aware design for human-aware AI, this talk explores my research and vision of next steps towards realizing adaptive closed-loop interactive autonomous systems that genuinely contemplate people.
- Speaker
- Richard Freedman
- Hosted by
- Felipe Meneguzzi
- Venue
- Meston 6