Journal article
An Ensemble Architecture for Learning Complex Problem-Solving Techniques from Demonstration
ACM transactions on intelligent systems and technology, v 3(4), pp 1-38
Sep 2012
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
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.
Metrics
Details
- Title
- An Ensemble Architecture for Learning Complex Problem-Solving Techniques from Demonstration
- Creators
- Xiaoqin Shelley Zhang - University of Massachusetts DartmouthBhavesh Shrestha - University of Massachusetts DartmouthSungwook Yoon - Arizona State UniversitySubbarao Kambhampati - Arizona State UniversityPhillip DiBona - Lockheed MartinJinhong K. Guo - Lockheed MartinDaniel McFarlane - Lockheed MartinMartin O. Hofmann - Lockheed MartinKenneth Whitebread - Lockheed MartinDarren Scott Appling - Georgia Institute of TechnologyElizabeth T. Whitaker - Georgia Institute of TechnologyEthan B. Trewhitt - Georgia Institute of TechnologyLi Ding - Rensselaer Polytechnic InstituteJames R. Michaelis - Rensselaer Polytechnic InstituteDeborah L. McGuinness - Rensselaer Polytechnic InstituteJames A. Hendler - Rensselaer Polytechnic InstituteJanardhan Rao Doppa - Oregon State UniversityCharles Parker - Oregon State UniversityThomas G. Dietterich - Oregon State UniversityPrasad Tadepalli - Oregon State UniversityWeng-Keen Wong - Oregon State UniversityDerek Green - Wyoming Department of EducationAnton Rebguns - Wyoming Department of EducationDiana Spears - Wyoming Department of EducationUgur Kuter - University of Maryland, College ParkGeoff Levine - University of Illinois SystemGerald Dejong - University of Illinois SystemReid L. MacTavish - Georgia Institute of TechnologySantiago Ontanon - Georgia Institute of TechnologyJainarayan Radhakrishnan - Georgia Institute of TechnologyAshwin Ram - Georgia Institute of TechnologyHala Mostafa - University of Massachusetts AmherstHuzaifa Zafar - University of Massachusetts AmherstChongjie Zhang - University of Massachusetts AmherstDaniel Corkill - University of Massachusetts AmherstVictor Lesser - University of Massachusetts AmherstZhexuan Song - Fujitsu
- Publication Details
- ACM transactions on intelligent systems and technology, v 3(4), pp 1-38
- Publisher
- Assoc Computing Machinery
- Number of pages
- 38
- Grant note
- FA8650-06-C-7605 / DARPA; United States Department of Defense; Defense Advanced Research Projects Agency (DARPA)
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Web of Science ID
- WOS:000313763500018
- Scopus ID
- 2-s2.0-84864766711
- Other Identifier
- 991021869110504721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Information Systems