Pat Langley
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Pat Langley has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
Daniel Shapiro
2 shared publications- Will Bridewell
Stanford University
2 shared publications Alan Fern
1 shared publicationAndrew Pohorille
Center for Computational Astrobiology and Fundamental Biology, NASA Ames Research Center, M/S 239-4, Moffett Field, CA 94035, USA
1 shared publicationDana Nau
1 shared publicationDavid J. Stracuzzi
1 shared publication
Latest Publications
- 1 January 2010
- Published by American Psychological Association (APA)
Transfer Learning in Integrated Cognitive Systems
- 1 January 2010
- Published by American Psychological Association (APA)
Transfer Learning in Integrated Cognitive Systems
- 1 November 2009
- Published by Oxford University Press (OUP)
- p. 216-234
Supporting Innovative Construction of Explanatory Scientific Models
Scientific modeling is a creative activity that can benefit from computational support. This chapter reports five challenges that arise in developing such aids, as illustrated by PROMETHEUS, a software environment that supports the construction and revision of explanatory models. These challenges include the paucity of relevant data, the need to incorporate prior knowledge, the importance of comprehensibility, an emphasis on explanation, and the practicality of user interaction. The responses to these challenges include the use of quantitative processes to encode models and background knowledge, as well as the combination of AND/OR search through a space of model structures with gradient descent to estimate parameters. This chapter reports our experiences with PROMETHEUS on three scientific modeling tasks and some lessons we have learned from those efforts. This chapter concludes by noting additional challenges that were not apparent at the outset of our work.
- 1 November 2009
- Published by Oxford University Press (OUP)
- p. 216-234
Supporting Innovative Construction of Explanatory Scientific Models
Scientific modeling is a creative activity that can benefit from computational support. This chapter reports five challenges that arise in developing such aids, as illustrated by PROMETHEUS, a software environment that supports the construction and revision of explanatory models. These challenges include the paucity of relevant data, the need to incorporate prior knowledge, the importance of comprehensibility, an emphasis on explanation, and the practicality of user interaction. The responses to these challenges include the use of quantitative processes to encode models and background knowledge, as well as the combination of AND/OR search through a space of model structures with gradient descent to estimate parameters. This chapter reports our experiences with PROMETHEUS on three scientific modeling tasks and some lessons we have learned from those efforts. This chapter concludes by noting additional challenges that were not apparent at the outset of our work.
- 30 September 2009
- Published by Elsevier in Cognitive Systems Research
- Vol. 10 (3) , 270-285
Skill transfer through goal-driven representation mapping
- 30 September 2009
- Published by Elsevier in Cognitive Systems Research
- Vol. 10 (3) , 270-285
Skill transfer through goal-driven representation mapping
- 11 December 2007
- Published by Springer Nature in Machine Learning
- Vol. 71 (1) , 1-32
Inductive process modeling
- 11 December 2007
- Published by Springer Nature in Machine Learning
- Vol. 71 (1) , 1-32
Inductive process modeling
- 16 June 2006
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 37 (3) , 191-201
Constructing explanatory process models from biological data and knowledge
- 16 June 2006
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 37 (3) , 191-201
Constructing explanatory process models from biological data and knowledge
- 1 March 2006
- Published by Elsevier in Ecological Modelling
- Vol. 194 (1-3) , 70-79
Inductive revision of quantitative process models
- 1 March 2006
- Published by Elsevier in Ecological Modelling
- Vol. 194 (1-3) , 70-79
Inductive revision of quantitative process models
- 1 October 2004
- Published by Springer Nature in Machine Learning
- Vol. 57 (1-2) , 13-34
Introduction: Lessons Learned from Data Mining Applications and Collaborative Problem Solving
- 1 October 2004
- Published by Springer Nature in Machine Learning
- Vol. 57 (1-2) , 13-34
Introduction: Lessons Learned from Data Mining Applications and Collaborative Problem Solving
- 1 April 2004
- Published by Defense Technical Information Center (DTIC)
Research on the Knowledge Plane: Distributed Learning and Reasoning About Network Outages
This report summarizes experiences developing a demonstration system to perform distributed detection of network failures. A simple system was developed based on probabilistically inferring failed nodes from observations on requested traffic and propagated error messages. Simulations verified that network failures could be automatically detected with limited memory and communication overhead.
- 1 April 2004
- Published by Defense Technical Information Center (DTIC)
Research on the Knowledge Plane: Distributed Learning and Reasoning About Network Outages
This report summarizes experiences developing a demonstration system to perform distributed detection of network failures. A simple system was developed based on probabilistically inferring failed nodes from observations on requested traffic and propagated error messages. Simulations verified that network failures could be automatically detected with limited memory and communication overhead.
- 25 August 1997
- Published by Defense Technical Information Center (DTIC)
Strong Autonomy for Physical Domains
A prototype of a strongly autonomous agent has been implemented. This prototype selects its own objectives and its own values. The agent can then calculate expected values, choose courses of action, and measure received reward.
- 25 August 1997
- Published by Defense Technical Information Center (DTIC)
Strong Autonomy for Physical Domains
A prototype of a strongly autonomous agent has been implemented. This prototype selects its own objectives and its own values. The agent can then calculate expected values, choose courses of action, and measure received reward.