Scott Niekum Postdoctoral Research Fellow The Robotics Institute, Carnegie Mellon University Contact Information Smith Hall, Room 228 Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 724.612.6191 sniekum@andrew.cmu.edu http://www.andrew.cmu.edu/user/sniekum Education 2009 2013 Doctor of Philosophy Computer Science Department, University of Massachusetts Amherst Dissertation: Semantically Grounded Learning from Unstructured Demonstrations Advisor: Professor Andrew G. Barto 2010 Master of Science Computer Science Department, University of Massachusetts Amherst 2001 2005 Bachelor of Science with Honors School of Computer Science, Carnegie Mellon University Additional major in Cognitive Science Honors Thesis: Reliable Rock Detection and Classification for Autonomous Science Research Experience 2013 present The Robotics Institute, Carnegie Mellon University Postdoctoral Research Fellow Mentor: Professor Christopher G. Atkeson Continuing research in robot learning from demonstration, Bayesian time-series analysis, and interactive perception. 2009 2013 The University of Massachusetts Amherst Research Assistant Performed research in robot learning from demonstration, Bayesian time-series analysis, and reinforcement learning topics. 2012 Willow Garage, Inc. Research Intern Mentor: Dr. Sachin Chitta Developed algorithms to learn multi-step tasks from unstructured demonstrations, such as assembling Ikea furniture. 2005 2009 Applied Perception, Inc. Robotics Engineer Developed software and hardware for autonomous ground vehicle projects, researched novel perception and planning techniques, and performed various on-site field tests. 2003 2005 The Robotics Institute, Carnegie Mellon University Research Assistant Mentor: Professor David Wettergreen Performed computer vision research for a NASA-funded project to further science autonomy capabilities on planetary rovers and performed field tests in the Atacama Desert.
Niekum 2/5 Teaching and Mentoring Experience Teaching Assistant: Fall 2009 and Spring 2010 Department of Computer Science, University of Massachusetts Amherst. I gave two 50-minute review lectures per week for an undergraduate computer science class in Java, Introduction to Problem Solving with Computers, in addition to holding weekly office hours and grading homework and exams. Ph.D. Student Intern: Karol Hausman (2014) I proposed and supervised Karol s research during his internship at Bosch Research under our shared NSF grant. This work resulted in submitting Active Articulation Model Estimation through Interactive Perception to ICRA 2015 and will contribute to Karol s Ph.D. dissertation at the University of Southern California. Ph.D. Student Intern: Russell Toris (2013) I helped to supervise Russell during his internship at Bosch Research under our shared NSF grant, resulting in the release of an open source ROS package for robot learning from demonstration. NSF REU: Nathaniel May (2010) I mentored Nathaniel as part of the NSF Research Experience for Undergraduates (REU) program at the University of Massachusetts Amherst, providing background knowledge and helping to guide his project. Grants and Awards NSF National Robotics Initiative Grant (NRI-Small) IIS-1208497: Multiple Task Learning from Unstructured Demonstrations ($499,911), 2012 2015 I was the primary author and researcher on this grant, though not able to serve as PI since I was a Ph.D. student at the time of writing. Through the grant s funding I performed and published independent research in addition to mentoring Ph.D. student interns at Bosch Research, our industry partner. University of Massachusetts Amherst Graduate School Fellowship ($16,000), 2011. Carnegie Mellon Alumni Award for best honors thesis ($1000), 2005. Invited Talks Brown University. Online Bayesian Changepoint Detection for Articulated Motion Models, October University of Michigan. Online Bayesian Changepoint Detection for Articulated Motion Models, September Georgia Institute of Technology. Online Bayesian Changepoint Detection for Articulated Motion Models, August University of Southern California. Grounded Learning from Unstructured Demonstrations, May Worcester Polytechnic Institute. Online Bayesian Changepoint Detection for Articulated Motion Models, May Carnegie Mellon University. Semantically Grounded Learning from Unstructured Demonstrations, April 2013. Massachusetts Institute of Technology. Semantically Grounded Learning from Unstructured Demonstrations, March 2013.
Niekum 3/5 Academic Service Organizing Co-organizer (with Andrea Thomaz and Sonia Chernova), AAAI 2015 Tutorial on Robot Learning from Demonstration, January 2015. Co-organizer (with Anca Dragan, James Boerkoel, and Sachin Chitta), RSS 2013 Workshop on Human-Robot Collaboration, June 2013. Journal Reviewing Autonomous Agents and Multi-Agent Systems (2014) Artificial Intelligence (2014) Frontiers in Computational Neuroscience (2013) IEEE Transactions on Autonomous Mental Development (2013) IEEE Transactions on Systems, Man, and Cybernetics (2012, 2013) Journal of Machine Learning Research (2011) Neurocomputing (2010) Conference Reviewing International Conference on Humanoid Robots (2014) Robotics: Science and Systems (2014) IEEE International Conference on Robotics and Automation (2014) International Conference on Intelligent Robots and Systems (2013, 2014) AAAI Special Robotics Track (2013) International Joint Conference on Artificial Intelligence (2013) AAAI Spring Symposium (2013) International Conference on Human Robot Interaction (2012, 2013) Neural Information Processing Systems (2011) North East Student Colloquium on Artificial Intelligence (2010) Publications Articles In Preparation or Under Review 1. S. Niekum, S. Osentoski, C.G. Atkeson, and A.G. Barto. Online Bayesian Changepoint Detection for Articulated Motion Models. (submitted to) IEEE International Conference on Robotics and Automation (ICRA), May 2015. 2. K. Hausman, S. Niekum, S. Osentoski, and G. Sukhatme. Active Articulation Model Estimation through Interactive Perception. (submitted to) IEEE International Conference on Robotics and Automation (ICRA), May 2015.
Niekum 4/5 Journal Articles 3. S. Niekum, S. Osentoski, G.D. Konidaris, S. Chitta, B. Marthi, and A.G. Barto. Learning Grounded Finite- State Representations from Unstructured Demonstrations. Accepted, International Journal of Robotics Research (IJRR), October 4. S. Niekum, A.G. Barto, L. Spector. Genetic Programming for Reward Function Search. IEEE Transactions on Autonomous Mental Development, vol.2, no.2, pp.83-90, June 2010. Highly Refereed Conference Publications 5. S. Niekum, S. Osentoski, S. Chitta, B. Marthi, and A.G. Barto. Incremental Semantically Grounded Learning from Demonstration. Robotics: Science and Systems (RSS), June 2013. 6. S. Niekum, S. Osentoski, G.D. Konidaris, and A.G. Barto. Learning and Generalization of Complex Tasks from Unstructured Demonstrations. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), October 2012. 7. S. Niekum and A.G. Barto. Clustering via Dirichlet Process Mixture Models for Portable Skill Discovery. Advances in Neural Information Processing Systems (NIPS), December 2011. 8. G.D. Konidaris, S. Niekum, and P.S. Thomas. TD γ : Reevaluating Complex Backups in Temporal Difference Learning. Advances in Neural Information Processing Systems (NIPS), December 2011. 9. D.R. Thompson, S. Niekum, T. Smith, and D. Wettergreen. Automatic Detection and Classification of Geological Features of Interest. IEEE Aerospace Conference, March 2005. 10. T. Smith, S. Niekum, D.R. Thompson, and D. Wettergreen. Concepts for Science Autonomy During Robotic Traverse and Survey. IEEE Aerospace Conference, March 2005. Lightly Refereed Workshops, Symposia, and Posters 11. S. Niekum, S. Osentoski, C.G. Atkeson, A.G. Barto. Learning Articulation Changepoint Models from Demonstration. RSS Workshop on Learning Plans with Context from Human Signals. July 12. G.D. Konidaris, S. Kuindersma, S. Niekum, R.A. Grupen and A.G. Barto. Robot Learning: Some Recent Examples. The Sixteenth Yale Workshop on Adaptive and Learning Systems, June 2013. 13. S. Niekum. An Integrated System for Learning Multi-Step Robotic Tasks from Unstructured Demonstrations. AAAI Spring Symposium: Reintegrating AI II, March 2013. 14. S. Niekum. Complex Task Learning from Unstructured Demonstrations. AAAI Doctoral Consortium, July 2012. 15. S. Niekum and A.G. Barto. Clustering via Dirichlet Process Mixture Models for Portable Skill Discovery. AAAI Workshop on Lifelong Learning from Sensorimotor Experience, August 2011. 16. S. Niekum, L. Spector, and A.G. Barto. Evolution of Reward Functions for Reinforcement Learning (poster abstract). Genetic and Evolutionary Computation Conference, June 2011. 17. S. Niekum. Evolved Intrinsic Reward Functions for Reinforcement Learning (extended abstract). Proceedings of the Twenty-Fourth Conference on Artificial Intelligence (AAAI), July 2010. Dissertations and Technical Reports 18. S. Niekum, S. Osentoski, C.G. Atkeson, A.G. Barto. CHAMP: Changepoint Detection Using Approximate Model Parameters. Technical report CMU-RI-TR-14-10, Robotics Institute, Carnegie Mellon University, June, 19. S. Niekum. Semantically Grounded Learning from Unstructured Demonstrations. Doctoral Dissertation, University of Massachusetts Amherst, September 2013. 20. S. Niekum. Reliable Rock Detection and Classification for Autonomous Science. Carnegie Mellon Senior Thesis, April 2005.
Niekum 5/5 Citizenship United States of America References 1. Professor Andrew G. Barto University of Massachusetts Amherst 140 Governors Drive, Amherst, MA 01003 +1 413-545-2109 barto@cs.umass.edu 2. Professor Christopher G. Atkeson Carnegie Mellon University 5000 Forbes Ave, Pittsbugh, PA 15213 +1 412-268-5544 cga@cs.cmu.edu 3. Professor Andrea L. Thomaz Georgia Institute of Technology 801 Atlantic Drive, Atlanta, GA 30332 +1 617-784-7154 athomaz@cc.gatech.edu 4. Dr. Sachin Chitta SRI International, Associate director of robotics systems and software 333 Ravenswood Avenue, Menlo Park, CA 94025 +1 650-859-2000 sachin.chitta@sri.com