Three members of the Interactive Robotics Group --- Matthew Gombolay, Brad Hayes, and Joseph Kim --- are presenting at the AAAI Conference on Artificial Intelligence (AAAI-17) in San Francisco, USA this week.
Add a splash of human intuition to planning algorithms! Joseph Kim will be presenting his paper at the planning track of the main conference on Monday, Feb 6th. Joseph will discuss improving automated planners by giving them the benefit of human high-level reasoning. You can read his paper titled, "Collaborative Planning with Encoding of Users' High-level Strategies" here.
Can machine learning be used to learn to teach? Gombolay et al. use a new algorithm to learn from teachers’ demonstration how to tutor students to solve a challenging military operations problem. Matthew's presentation will be on Satruday, February 4th at the Human-Machine Collaborative Learning workshop.
Brad Hayes, along with co-presenters Ece Kamar and Matthew Taylor, will conduct the "Interactive Machine Learning: From Classifiers to Robotics" tutorial on February 5th (link: http://interactiveml.net). The tutorial will cover a range of topics including crowdsourcing algorithms, sequential task learning from demonstration, and human-in-the-loop reinforcement learning.