Learning from Demonstration is an effective method for interactively teaching skills to a robot learner. However, a skill learned via demonstrations is often learned within a particular environment and uses a specific set of objects, and thus may not be immediately applicable for use in unfamiliar environments. Transfer learning addresses this problem by enabling a robot to apply learned skills to unfamiliar environments. We describe our ongoing work to develop a system which enables transfer learning by representing skill demonstrations according to the level of similarity between the source and target environments.
Recent News
New Podcast on Jill Watson and SAMI
Hiring a full-time research scientist and a half-time post-doc
News coverage on Jill Watson: what different sectors can teach us about AI
Congratulations to DILab alumni Mukundan Kuthalam for his recent acceptance to the Computer Science PhD program at Northwestern University!
Congratulations to DILab alumni Varsha Achar for starting her new job at Facebook!