The data collected so far is suitable for comparing classification algorithms - we are interested in extending the data here so that reinforcement learning, and other active robot control learning strategies could be compared on real world data. Any comments or suggestions would be appreciated.
The following paper for more details on why this was created, and how it is used:
Joseph O'Sullivan. ``The CMU Learning Robot Laboratory Data Toolkit''. In Proceedings of the MLC-Colt '94 Workshop on Robot Learning . Rutgers, The State University of New Jersey, New Brunswick, July 9, 1994.Last Updated: 2Nov94 19:00 email@example.com