13 March 2015 |
The TERESA partners are organising a Workshop on Machine Learning for Social Robotics at ICRA-15 in Seattle, Washington on May 26th 2015.
The workshop considers the unique challenges and opportunities that arise in developing machine learning techniques for social robotics. These challenges include learning in the presence of humans and learning with humans in the loop, which can involve imitation learning, co-learning, or inverse reinforcement learning. Machine learning challenges can also focus on perceiving something about the humans in the robot’s environment or behaving in a socially appropriate way. In all cases, learning must cope with enormous variation in human behavior and evaluation of robot behavior, which make learning especially difficult.