|
Matthew Klenk klenk.matt@gmail.com Postdoctoral Research Associate Navy Center for Applied
Research in Artificial Intelligence Washington,
DC |
Education B.A. Computer Science,
Emory University, May 2003 Research Interests In my research, I seek to
understand human reasoning by building computational models of it. This goal has led to three areas or
research: analogical reasoning and learning, intelligent architecture
development, and cognitive modeling.
One application of particular interest to me is that of a
collaborative assistant. I recently
wrote an article for Forbes
on the subject. Analogical Reasoning and Learning Analogy has long been seen as a method for knowledge based systems to overcome brittleness. But there are still many open questions concerning exactly when and how analogy can be employed to overcome this issue. My dissertation, Using Analogy to Overcome Brittleness in AI Systems, introduced the following two analogical methods.
Relevent Papers: Klenk, M. and Forbus, K. 2007. Measuring the level of transfer learning by an AP Physics problem-solver. Proceedings of AAAI-07: 22nd National Conference on Artificial Intelligence, Vancouver, BC. [pdf] Klenk, M., Forbus, K., Tomai, E., Kim,H., and Kyckelhahn, B. 2005. Solving Everyday Physical Reasoning Problems by Analogy using Sketches. Proceedings of AAAI-05: 20th National Conference on Artificial Intelligence, Pittsburgh, USA. [pdf] Intelligent Architecture Development When we begin to think of AI systems that exist over time operating over a range of tasks requiring intelligence, a number of important research areas arise. I have begun to addressing some of these while working on the Companions cognitive systems project.
Relevant Papers: Forbus, K., Klenk, M., and Hinrichs, T. 2009. Companion Cognition Systems: Design Goals and Some Lessons Learned. IEEE-Intelligent Systems, Special Issue on “Human-level Intelligence” Klenk, M. 2009. Transfer as a Benchmark for Multi-Representational Architectures. AAAI Fall Symposium on Multi-Representational Architectures, Washington, DC. [pdf] Cognitive Modeling In my research, I seek to understand human reasoning by building computational models of it. This is necessarily an interdisciplinary project. The analogical reasoning methods from my thesis build upon existing Cognitive Science research of human analogical retrieval, matching, and problem-solving. In addition, I am also interested in other aspects of higher-order cognition resulting in the following models:
Relevant Papers: Paritosh, P.K. and Klenk, M. 2006. Cognitive Processes in Quantitative Estimation: Analogical Anchors and Causal Adjustment. Proceedings of the 28th Annual Conference of the Cognitive Science Society (CogSci). Vancouver, BC. [pdf] Dehghani, M., Tomai, E.,
Forbus, K., Iliev, R., and Klenk, M. 2008. MoralDM: A Computational
Modal of Moral Decision-Making. Proceedings of the 30th Annual Conference
of the Cognitive Science Society (CogSci), Washington, D.C. [pdf] Publications [cv] Dissertation
1. Klenk, M. 2009. Using Analogy to Overcome Brittleness in AI Systems. Department of Electrical Engineering and Computer Science. Northwestern University. June 2009. [pdf] Selected Conferences and Journals
1. Klenk, M. and Forbus, K. (under review). Analogical Model Formulation for Transfer Learning in AP Physics. Artificial Intelligence. Elsevier. 2. Klenk, M., Forbus, K., Tomai, E., and Kim, H. (under review). Using Analogical Model Formulation with Sketches to Solve Bennett Mechanical Comprehension Test Problems. Journal Experimental and Theoretical Artificial Intelligence, Special Issue on “Test-Based AI”. Taylor & Francis. 3. Klenk, M. and Forbus, K. 2009. Domain Transfer via Cross-Domain Analogy. Cognitive Systems Research, Special Issue on “Analogies: Integrating Cognitive Abilities”. Elsevier. [pdf] 4. Forbus, K., Klenk, M., and Hinrichs, T. 2009. Companion Cognition Systems: Design Goals and Some Lessons Learned. IEEE-Intelligent Systems, Special Issue on “Human-level Intelligence” 5. Dehghani, M., Tomai, E., Forbus, K., and Klenk, M. 2008. An Integrated Reasoning Approach to Moral Decision-Making. Proceedings of AAAI-08: 23rd National Conference on Artificial Intelligence. Chicago, IL. 24% acceptance rate. [pdf] 6. Klenk, M. and Forbus, K. 2007. Measuring the level of transfer learning by an AP Physics problem-solver. Proceedings of AAAI-07: 22nd National Conference on Artificial Intelligence, Vancouver, BC. 27% acceptance rate. [pdf] 7. Klenk, M., Forbus, K., Tomai, E., Kim,H., and Kyckelhahn, B. 2005. Solving Everyday Physical Reasoning Problems by Analogy using Sketches. Proceedings of AAAI-05: 20th National Conference on Artificial Intelligence, Pittsburgh, USA. 18% acceptance rate. [pdf] Conferences
1. Klenk, M. and Forbus, K. 2009. Persistent Mappings in Cross-Domain Analogical Learning of Physics Domains. Proceedings of the 2nd International Analogy Conference. Sofia, Bulgaria. [pdf] 2. Dehghani, M., Tomai, E., Forbus, K., Iliev, R., and Klenk, M. 2008. MoralDM: A Computational Modal of Moral Decision-Making. Proceedings of the 30th Annual Conference of the Cognitive Science Society (CogSci-08), Washington, D.C. [pdf] 3. Klenk, M. and Forbus, K. 2007. Cognitive modeling of analogy events in physics problem solving from examples. Proceedings of the 29th Annual Conference of the Cognitive Science Society (CogSci-07). Nashville, TN. [pdf] 4. Paritosh, P.K. and Klenk, M. 2006. Cognitive Processes in Quantitative Estimation: Analogical Anchors and Causal Adjustment. Proceedings of the 28th Annual Conference of the Cognitive Science Society (CogSci-06). Vancouver, Canada. [pdf] Workshops and Symposia
1. Klenk, M. 2009. Transfer as a Benchmark for Multi-Representational Architectures. AAAI Fall Symposium on Multi-Representational Architectures, Washington, DC. [pdf] 2. Laviers, K., Sukthankar, G., Klenk, M., Aha, D., and Molineaux, M. 2009. Opponent Modeling and Spatial Similarity to Retrieve and Reuse Superior Plays. ICCBR Workshop on Case-Based Reasoning for Computer Games. Seattle, WA. [pdf] 3. Forbus, K., Hinrichs, T., and Klenk, M., 2008. Companion Cognitive Systems: Design Goals and Some Lessons Learned. AAAI Fall Symposium on Naturally-Inspired Artificial Intelligence, Washington, DC. [pdf] 4. Dehghani, M., Tomai, E., Forbus, K., Iliev, R., and Klenk, M. 2008. MoralDM: A Computational Modal of Moral Decision-Making. Abstract accepted at the 2008 meeting of Society of Judge and Decision Making (SJDM). Chicago, IL. 5. Klenk, M., Friedman, S., and Forbus, K. 2008. Learning Modeling Abstractions via Generalization. 22nd International Workshop on Qualitative Reasoning. Boulder, CO. [pdf] 6. Dehghani, M., Tomai, E., Forbus, K., and Klenk, M. 2008. Order of Magnitude Reasoning in Modeling Moral Decision-Making. 22nd International Workshop on Qualitative Reasoning. Boulder, CO. [pdf] 7. Klenk, M. and Forbus, K. 2007. Cross domain analogies for learning domain theories. In Angela Schwering et al. (Eds.), Analogies: Integrating Multiple Cognitive Abilities. Publications of the Institute of Cognitive Science, University of Osnabrück, Volume 5-2007 [pdf] 8. Klenk, M. and Forbus, K. 2007. Learning domain theories via analogical transfer. Proceedings of 21st International Workshop on Qualitative Reasoning Workshop. Aberystwyth, U.K. [pdf] 9. Klenk, M. and Forbus, K. 2006. Analogical Model Formulation for AP Physics Problems. 20th International Workshop on Qualitative Reasoning. Hanover, USA. [pdf] 10. Klenk, M., Forbus, K., Tomai, E., Kim,H., and Kyckelhahn, B. 2005. Solving Everyday Physical Reasoning Problems by Analogy using Sketches. Proceedings of 19th International Workshop on Qualitative Reasoning. Graz, Austria. [pdf] 11. Forbus, K., Lockwood, K., Klenk, M., Tomai, E., and Usher, J. 2004. Open-domain sketch understanding: The nuSketch approach. AAAI Fall Symposium on Making Pen-based Interaction Intelligent and Natural, Washington, DC, USA. [pdf] Awards
|
|