TY - JOUR
T1 - An ensemble architecture for learning complex problem-solving techniques from demonstration
AU - Zhang, Xiaoqin Shelley
AU - Shrestha, Bhavesh
AU - Yoon, Sungwook
AU - Kambhampati, Subbarao
AU - Dibona, Phillip
AU - Guo, Jinhong K.
AU - McFarlane, Daniel
AU - Hofmann, Martin O.
AU - Whitebread, Kenneth
AU - Appling, Darren Scott
AU - Whitaker, Elizabeth T.
AU - Trewhitt, Ethan B.
AU - Ding, Li
AU - Michaelis, James R.
AU - McGuinness, Deborah L.
AU - Hendler, James A.
AU - Doppa, Janardhan Rao
AU - Parker, Charles
AU - Dietterich, Thomas G.
AU - Tadepalli, Prasad
AU - Wong, Weng Keen
AU - Green, Derek
AU - Rebguns, Anton
AU - Spears, Diana
AU - Kuter, Ugur
AU - Levine, Geoff
AU - Dejong, Gerald
AU - MacTavish, Reid L.
AU - Ontañón, Santiago
AU - Radhakrishnan, Jainarayan
AU - Ram, Ashwin
AU - Mostafa, Hala
AU - Zafar, Huzaifa
AU - Zhang, Chongjie
AU - Corkill, Daniel
AU - Lesser, Victor
AU - Song, Zhexuan
PY - 2012/9
Y1 - 2012/9
N2 - We present a novel ensemble architecture for learning problem-solving techniques from a very small number of expert solutions and demonstrate its effectiveness in a complex real-world domain. The key feature of our "Generalized Integrated Learning Architecture" (GILA) is a set of heterogeneous independent learning and reasoning (ILR) components, coordinated by a central meta-reasoning executive (MRE). The ILRs are weakly coupled in the sense that all coordination during learning and performance happens through the MRE. Each ILR learns independently from a small number of expert demonstrations of a complex task. During performance, each ILR proposes partial solutions to subproblems posed by the MRE, which are then selected from and pieced together by the MRE to produce a complete solution. The heterogeneity of the learner-reasoners allows both learning and problem solving to be more effective because their abilities and biases are complementary and synergistic. We describe the application of this novel learning and problem solving architecture to the domain of airspace management, where multiple requests for the use of airspaces need to be deconflicted, reconciled, and managed automatically. Formal evaluations show that our system performs as well as or better than humans after learning from the same training data. Furthermore, GILA outperforms any individual ILR run in isolation, thus demonstrating the power of the ensemble architecture for learning and problem solving.
AB - We present a novel ensemble architecture for learning problem-solving techniques from a very small number of expert solutions and demonstrate its effectiveness in a complex real-world domain. The key feature of our "Generalized Integrated Learning Architecture" (GILA) is a set of heterogeneous independent learning and reasoning (ILR) components, coordinated by a central meta-reasoning executive (MRE). The ILRs are weakly coupled in the sense that all coordination during learning and performance happens through the MRE. Each ILR learns independently from a small number of expert demonstrations of a complex task. During performance, each ILR proposes partial solutions to subproblems posed by the MRE, which are then selected from and pieced together by the MRE to produce a complete solution. The heterogeneity of the learner-reasoners allows both learning and problem solving to be more effective because their abilities and biases are complementary and synergistic. We describe the application of this novel learning and problem solving architecture to the domain of airspace management, where multiple requests for the use of airspaces need to be deconflicted, reconciled, and managed automatically. Formal evaluations show that our system performs as well as or better than humans after learning from the same training data. Furthermore, GILA outperforms any individual ILR run in isolation, thus demonstrating the power of the ensemble architecture for learning and problem solving.
KW - Complex problemsolving
KW - Ensemble architecture
KW - Learning from demonstration
UR - http://www.scopus.com/inward/record.url?scp=84864766711&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84864766711&partnerID=8YFLogxK
U2 - 10.1145/2337542.2337560
DO - 10.1145/2337542.2337560
M3 - Article
AN - SCOPUS:84864766711
SN - 2157-6904
VL - 3
JO - ACM Transactions on Intelligent Systems and Technology
JF - ACM Transactions on Intelligent Systems and Technology
IS - 4
M1 - 75
ER -