An Ensemble Learning and Problem Solving Architecture for Airspace Management

Printer-friendly version

Abstract:

In this paper we describe the application of a novel learning and problem solving architecture to the domain of airspace management, where multiple requests for the use of airspace need to be reconciled and managed automatically. The key feature of our "Generalized Integrated Learning Architecture" (GILA) is a set of integrated learning and reasoning (ILR) systems coordinated by a central meta-reasoning executive (MRE). Each ILR learns independently from the same training example and contributes to problem-solving in concert with other ILRs as directed by the MRE. Formal evaluations show that our system performs as well as or better than humans after learning from the same training data. Further, GILA outperforms any individual ILR run in isolation, thus demonstrating the power of the ensemble architecture for learning and problem solving.

History

DateCreated ByLink
July 18, 2011
15:06:01
Patrick WestDownload

Related Projects:

Generalized Integrated Learning Architecture (GILA)
Principal Investigator: Jim Hendler and Deborah L. McGuinness
Description: The Generalized Integrated Learning Architecture [GILA] is a general-purpose integrated multi-agent platform that solves domain problems by learning from a problem-solution pair submitted by a human expert. One of the key purposes of GILA is to learn how humans solve complex problems and apply this knowledge to future problems. A complex problem domain known as the Airspace Control Scenario has been chosen to drive the development of GILA and evaluate its performance. The objective of this problem domain is to resolve conflicts in a collection of airspace allocations for aircrafts.
Inference Web Project LogoInference Web
Principal Investigator: Deborah L. McGuinness
Description: The Inference Web is a Semantic Web based knowledge provenance infrastructure that supports interoperable explanations of sources, assumptions, learned information, and answers as an enabler for trust. Provenance - if users (humans and agents) are to use and integrate data from unknown, uncertain, or multiple sources, they need provenance metadata for evaluation Interoperability - more systems are using varied sources and multiple information manipulation engines, thus increasing interoperability requirements Explanation/Justification - if information has been manipulated (i.e., by sound deduction or by heuristic processes), information manipulation trace information should be available Trust - if some sources are more trustworthy than others, trust ratings are desired The Inference Web consists of two important components: Proof Markup Language (PML) Ontology - Semantic Web based representation for exchanging explanations including provenance information - annotating the sources of knowledge justification information - annotating the steps for deriving the conclusions or executing workflows trust information - annotating trustworthiness assertions about knowledge and sources IW Toolkit - Web-based and standalone tools that facilitate human users to browse, debug, explain, and abstract the knowledge encoded in PML.

Related Research Areas:

Data Frameworks
Lead Professor: Peter Fox
Description: None.
Concepts:
Knowledge Provenance
Lead Professor: Deborah L. McGuinness
Description: Knowledge Provenance
Concepts: Provenance,