Curriculum Vitae
Education
University of Maryland, Baltimore CountyB.S., Computer Science 2007
Research Interests
I am interested in a wide range of artificial intelligence research, but I primarily focus on reinforcement learning, autonomous planning, and human-agent interaction. More specifically, I have published research in learning from human-delivered reinforcement, learning from demonstration, natural language command grounding, multi-agent interaction, hierarchical action learning, and learning planning knowledge. I have contributed a large breadth of open source research tools for these topics as the creator the Brown-UMBC Reinforcement Learning and Planning Java library.
Employment
Adjunct Research Professor | March 2016-Present |
Brown University | Providence, RI |
Postdoctoral Researcher | March 2013-March 2016 |
Brown University | Providence, RI |
Instructor | Fall 2009, 2010, 2011 |
University of Maryland Baltimore County | Baltimore, MD |
Courses: Computer science 104: Problem Solving and Computer Programming; Computer Science 100: Introduction to Computer Science; TRS 201: Introduction to Unix and Linux for transfer students |
Significant Software Projects
BURLAPBURLAP is a large open-source reinforcement learning and planning Java library that I created. The library supports a wide range of different single and multi-agent problems. Algorithms included range from classic forward search planning, to value function approximation, to inverse reinforcement learning for learning from humans. Many standard domains are included and the library also includes a range of analysis and visualization tools. The library and tutorials can be found at http://burlap.cs.brown.edu.
Java Rosbridge
Java Rosbridge is a Java library for connecting Java code to Robot OS (ROS) over Rosbridge, thereby allowing arbitrary Java code run on local or remote machines to control robots that interface with ROS. The library is available at https://github.com/h2r/java_rosbridge.
BurlapCraft
BurlapCraft is a mod for the video game Minecraft that allows researchers to run AI algorithms within Minecraft, using BURLAP as the interface for implementing AI algorithms. I created BurlapCraft in conjunction with Krishna Aluru, Stefanie Tellex, and John Oberlin. BurlapCraft creates well defined state representations of the game world and provides action controllers to manipulate the Minecraft player. Along with a provided model of the world, these tools are bundled together to allow existing planning and learning algorithms developed in BURLAP to control the Minecraft player. The library is available at https://github.com/h2r/burlapcraft.
Major League Kickball 2010
This is a sports game developed for the iPhone and iPod Touch. It was available on the iTunes App Store and was released on November 11th, 2010. More information available from http://www.onesouthdesign.com/apps.html
Berserker
This is an action puzzle game developed for the iPhone and iPod Touch. It was available on the iTunes App Store and was released on December 23rd, 2009. More information available from http://www.onesouthdesign.com/apps.html
QuickStat
This is a fast light weight statistical program for the iPhone/iPod Touch. Users are able to quickly enter data for 2 variables and various statistical tests are computed for the data in real time. This program was released for sale at the iTunes App Store on March 4th, 2009.
Primer Match
Program to assist in localizing forward and reverse DNA/RNA primer matches in large DNA sequences.
IRB/Regulatory Deadline Database
A database for the Division of Allergy and Clinical Immunology at Johns Hopkins to track various regulatory deadlines for human subjects research projects.
Publications
- N. Gopalan, M. desJardins, M. L. Littman, J. MacGlashan, S. Squire, S. Tellex, J. Winder, and L. L. S. Wong. Planning with Abstract Markov Decision Processes. In ICML workshop on Abstraction in Reinforcement Learning, 2016.
- B. Peng, J. MacGlashan, R. Loftin, M. L. Littman, D. L. Roberts, and M. E. Taylor. A Need for Speed: Adapting Agent Action Speed to Improve Task Learning from Non-Expert Humans. In Proceedings of the 2016 International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2016.
- B. Peng, J. MacGlashan, R. Loftin, M. L. Littman, D. L. Roberts, and M. E. Taylor. An Empirical Study of Non-Expert Curriculum Design for Machine Learners. In Proceedings of the Interactive Machine Learning workshop (at IJCAI), New York City, NY, USA, July 2016.
- K. Aluru, S. Tellex, J. Oberlin, and J. MacGlashan, "Minecraft as an experimental world for AI in robotics," In AAAI Fall Symposium, 2015.
- E. Wu, Y. Han, D. Whitney, J. Oberlin, J. MacGlashan, and S. Tellex, “Robotic social feedback for object specification,” in AAAI Fall Symposium, 2015.
- E.D. Hershkowitz, J. MacGlashan, and S. Tellex, "Learning propositional functions for planning and reinforcement learning," In AAAI Spring Symposium, 2015.
- J. MacGlashan, M. Babes-Vroman, M. desJardins, M. Littman, S. Muresan, S. Squire, S. Tellex, D. Arumugam, and L. Yang, "Grounding English commands to reward functions," in Robotics: Science and Systems, 2015.
- J. MacGlashan and M. L. Littman, "Between Imitation and Intention Learning," in Proceedings of the International Joint Conference on Artificial Intelligence, 2015.
- N. Topin, N. Haltmeyer, S. Squire, J. Winder, M. desJardins, and J. MacGlashan, "Portable option discovery for automated learning transfer in object-oriented markov decision processes," in Proceedings of the International Joint Conference on Artificial Intelligence, 2015.
- D. Abel, D. Hershkowitz, G. Barth-Maron, S. Brawner, K. O’Farrell, J. MacGlashan, and S. Tellex, "Goal-based action priors," in The International Conference on Automated Planning and Scheduling, 2015.
- R. Loftin, B. Peng, J. MacGlashan, M. L. Littman, M. E. Taylor, J. Huang, and D. L. Roberts, "Learning behaviors via human-delivered discrete feedback: modeling implicit feedback strategies to speed up learning," Autonomous Agents and Multi-Agent Systems, pp. 1–30, 2015.
- Loftin, R., MacGlashan, J., Peng, B., Taylor, M., Littman, M., Huang, J., Roberts, D., "A Strategy-Aware Technique for Learning Behaviors from Discrete Human Feedback" in Proceedings of the AAAI Conference, 2014.
- Loftin, R., Peng, B., MacGlashan, J., Littman, M., Taylor, M., Huang, J., Roberts, D., "Learning Something from Nothing: Leveraging Implicit Human Feedback Strategies" in Proceedings of the 23rd IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN '14), 2014.
- MacGlashan, J., Littman, M., Loftin, R., Peng, B., Roberts, D., Taylor, M., "Training an Agent to Ground Commands with Reward and Punishment," in Proceedings of the AAAI Machine Learning for Interactive Systems Workshop, 2014.
- Abel, D., Barth-Maron, G., MacGlashan, J., Tellex, S., "Towards Affordance-aware Planning" in Proceedings of the RSS Affordances in Vision for Cognitive Robotics Workshop, 2014.
- MacGlashan, J., Littman, M., Babes-Vroman, M., Squire, S., desJardins, M., and Muresan, S., "Translating English to Reward Functions," Computer Science Department Brown University Tech. Rep. CS-14-01 2014.
- James MacGlashan, Monica Babes-Vroman, Kevin Winner, Ruoyuan Gao, Richard Adjogah, Marie desJardins, Michael Littman, Smaranda Muresan, "Learning to Interpret Natural Language Instructions" in Proceedings of the AAAI Workshop on Grounding Language for Physical Systems, 2012.
- James MacGlashan, Don Miner, Marie desJardins,"A Game Playing System for Use in Computer Science Education," in Proceedings of the 23rd Florida Artificial Intelligence Research Society Conference, May 2010.
- Marie desJardins, James MacGlashan, and Kiri Wagstaff, "Confidence-Based Feature Acquisition to Minimize Training and Test Costs," in Proceedings of the 2010 SIAM International Conference on Data Mining, April 2010.
- James MacGlashan, Marie desJardins, "Hierarchical Skill Learning for High-level Planning," in Proceedings of the ICML/UAI/COLT Workshop on Abstraction in Reinforcement Learning, Montreal, Canada, 2009.
Other versions appeared in:
- 2009 Doctoral Consortium for the International Conference on Automated Planning and Scheduling
- 2010 Doctoral Consortium for the The AAAI Conference on Artificial Intelligence
- Marie desJardins, James MacGlashan, and Julia Ferraioli, "Interactive visual clustering for relational data," in Constrained Clustering Advances in Algorithms Theory and Applications, Kiri Wagstaff, Sugato Basu, Ian Davidson, Ed., chapter 14, pp. 329-355. Chapman & Hall/CRC Press, Boca Raton, 2008
- Marie desJardins, James MacGlashan, and Julia Ferraioli, "Interactive visual clustering," Proceedings of the 2007 International Conference on Intelligent User Interfaces, January 2007
Scholarships and Awards
- ICML 2009 Student Scholarship
- ICAPS 2009 Doctorial Consoritum and Scholarship
- PROMISE 2010 Rocky Gap Dissertation House Retreate
- SDM 2010 Doctoral Consoritium and Scholarship
- AAAI 2010 Doctoral Consoritium and Scholarship