Filters








19,983 Hits in 11.7 sec

Exact and Approximate Weighted Model Integration with Probability Density Functions Using Knowledge Compilation

Pedro Zuidberg Dos Martires, Anton Dries, Luc De Raedt
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
We show how standard knowledge compilation techniques (to SDDs and d-DNNFs) apply to weighted model integration, and use it in two novel solvers, one exact and one approximate solver.  ...  Weighted model counting has recently been extended to weighted model integration, which can be used to solve hybrid probabilistic reasoning problems.  ...  for weighted model integration and knowledge compilation.  ... 
doi:10.1609/aaai.v33i01.33017825 fatcat:vdccjb53qbdjtkvle6iytrpb2e

Integrating Symbol-Oriented and Sub-Symbolic Reasoning Methods into Hybrid Systems [chapter]

Franz J. Kurfess
2002 From Synapses to Rules  
This contribution discusses approaches to integrate symbol-oriented reasoning methods with sub-symbolic ones into hybrid systems.  ...  Knowledge representation and reasoning methods in artificial intelligence almost exclusively rely on symbol-oriented methods: Statements describing aspects and objects of the system to be modelled are  ...  INTEGRATION OF SYMBOL-ORIENTED AND SUB-SYMBOLIC SYSTEMS A combination of different approaches to knowledge representation and processing, with expert systems as example of symbol-oriented system and neural  ... 
doi:10.1007/978-1-4615-0705-5_14 fatcat:yivvfaup5zbdpcrau5uxenfoo4

A General Outpatient Triage System Based on Dynamic Uncertain Causality Graph

Xusong Bu, Lin Lu, Zhan Zhang, Qin Zhang, Yan Zhu
2020 IEEE Access  
To validate the triage system, we constructed a basic triage knowledge base and adjusted it to match the triage requirement of a county-level hospital.  ...  The hybrid DUCG is a combination of the single-valued DUCG (S-DUCG) and the multivalued DUCG (M-DUCG). In this paper, the M-DUCG is utilized to construct the basic outpatient triage knowledge base.  ...  Medicine of the target hospital, and BX 7 is used as the symbol of the Respiratory Medicine in the hybrid DUCG.  ... 
doi:10.1109/access.2020.2995087 fatcat:wpiurmoyz5duxa5t42uohn2lja

Symbolic Logic meets Machine Learning: A Brief Survey in Infinite Domains [article]

Vaishak Belle
2020 arXiv   pre-print
The deduction camp concerns itself with questions about the expressiveness of formal languages for capturing knowledge about the world, together with proof systems for reasoning from such knowledge bases  ...  In AI, historically, these camps have loosely divided the development of the field, but advances in cross-over areas such as statistical relational learning, neuro-symbolic systems, and high-level control  ...  See [24] for a comprehensive view on the use of knowledge compilation for machine learning.  ... 
arXiv:2006.08480v1 fatcat:d23e4d6kcfbtjduztfr6y4536a

Scaling up Probabilistic Inference in Linear and Non-linear Hybrid Domains by Leveraging Knowledge Compilation

Anton Fuxjaeger, Vaishak Belle
2020 Proceedings of the 12th International Conference on Agents and Artificial Intelligence  
In this work, we propose a new implementation regime that leverages propositional knowledge compilation for scaling up inference.  ...  In particular, we use sentential decision diagrams, a tractable representation of Boolean functions, as the underlying model counting and model enumeration scheme.  ...  of Informatics, University of Edinburgh.  ... 
doi:10.5220/0008896003470355 dblp:conf/icaart/FuxjaegerB20 fatcat:xcqswmbd3zairadnl4cbzy6itq

Scaling up Probabilistic Inference in Linear and Non-Linear Hybrid Domains by Leveraging Knowledge Compilation [article]

Anton Fuxjaeger, Vaishak Belle
2020 arXiv   pre-print
In this work, we propose a new implementation regime that leverages propositional knowledge compilation for scaling up inference.  ...  In particular, we use sentential decision diagrams, a tractable representation of Boolean functions, as the underlying model counting and model enumeration scheme.  ...  of Informatics, University of Edinburgh.  ... 
arXiv:1811.12127v2 fatcat:4ntcctbcvnfgnmtc5bgwtovgdi

Neurosymbolic AI: The 3rd Wave [article]

Artur d'Avila Garcez, Luis C. Lamb
2020 arXiv   pre-print
We focus on research that integrates in a principled way neural network-based learning with symbolic knowledge representation and logical reasoning.  ...  Many have identified the need for well-founded knowledge representation and reasoning to be integrated with deep learning and for sound explainability.  ...  on hybrid neural-symbolic systems, we are acutely aware of the tension between principled integration and practical value and application.  ... 
arXiv:2012.05876v2 fatcat:m5g4kvcvo5epzmlsmft7vj5hva

Development of knowledge-based systems in wind engineering

Ahsan Kareem, Robert H. Allen
1990 Journal of Wind Engineering and Industrial Aerodynamics  
We present an example of an expert system under development. Specifically, object-oriented and rule-based representations are used to encode knowledge about the serviceability of high-rise buildings.  ...  In addition to this, the development market of expert systems has burgeoned with both specialized hardware (e.g., Sy.mbolics and Lambda computers) and software (e.g., OPS5, KEE TM, and ARTrM).  ...  ACKNOWLEDGMENTS Financial assistance provided by the National Science Foundation Grant ECE-8352223 and American Institute of Steel Construction is gratefully acknowledged.  ... 
doi:10.1016/0167-6105(90)90121-r fatcat:rjqz67j7qzbuxkqk3ttewy3haa

Development of knowledge-based systems in wind engineering

A Kareem
1990 Journal of Wind Engineering and Industrial Aerodynamics  
We present an example of an expert system under development. Specifically, object-oriented and rule-based representations are used to encode knowledge about the serviceability of high-rise buildings.  ...  In addition to this, the development market of expert systems has burgeoned with both specialized hardware (e.g., Sy.mbolics and Lambda computers) and software (e.g., OPS5, KEE TM, and ARTrM).  ...  ACKNOWLEDGMENTS Financial assistance provided by the National Science Foundation Grant ECE-8352223 and American Institute of Steel Construction is gratefully acknowledged.  ... 
doi:10.1016/0167-6105(90)90466-p fatcat:fhy7b4xugzah3nt4p6x7324qtm

Thoughts and Afterthoughts on the 1988 Workshop on Principles of Hybrid Reasoning

Alan M. Frisch, Anthony G. Cohn
1991 The AI Magazine  
of human memory and learning.  ...  His current interests are relating AI programming to traditional scientific modeling, studying computer systems in the workplace, and re-examining the relation of cognitive science theories to the processes  ...  Acknowledgements We thank the AAAI for their financial support of this workshop and Claudia Mazzetti and the AAAI staff for their assistance with arranging the workshop.  ... 
doi:10.1609/aimag.v11i4.876 dblp:journals/aim/FrischC91 fatcat:z5cmhgjqg5fetnrzuqsfqfsmiq

Symbolic and Sub-Symbolic Representations in Computational Models of Human Cognition

Troy D. Kelley
2003 Theory & psychology  
The argument is this: Does the human cognitive system use symbols as a representation of knowledge, and does it process symbols and their respective constituents?  ...  Or does the human cognitive system use a distributed representation of knowledge, and is it somehow capable of processing this distributed representation of knowledge in a complex and meaningful way?  ...  Does the human cognitive system use symbols as a representation of knowledge, and process symbols and their respective constituents?  ... 
doi:10.1177/0959354303136005 fatcat:rvi5rienu5acrcdprk5dpddwxq

CoLab: A hybrid knowledge representation and compilation laboratory

Harold Boley, Philipp Hanschke, Knut Hinkelmann, Manfred Meyer
1995 Annals of Operations Research  
It will be used as the source of examples throughout the paper when discussing t he enhanced formalisms, the hybrid representation architecture, and the compilers.  ...  -reasoning should be integrated with backward (top-down) reasoning. Constraint propagation should be used wherever possible instead of search-intensive resolution.  ...  Acknow ledgements The research presented in this paper and the development of COLAB have been carried out by the knowledge-compilation group of the ARC-TEC project at the DFKI, supported by the BMFT under  ... 
doi:10.1007/bf02031716 fatcat:djzxr64dgvce5bdiug7sdfgoca

Subject index to volume 2

1988 International Journal of Approximate Reasoning  
, construction of, 337 Bayes' rule of conditioning, 328 Bayesian analysis, decision tree induction system and, 330 Bayesian approach, heuristic, to knowledge acquisition, application to analysis of tissuetype  ...  uncertainty and temporal relationships in, 337 Autonomous agents, centrality of, in theories of action under uncertainty, 303-326 Backward chaining, with fuzzy goals and rules, 108 Bayes belief network  ...  systems, 273-278 Knowledge networks, combined and adaptive, hidden patterns in, 377-393 Learning concepts, logical aspects of, 349-364 in knowledge-based systems, with imperfect teacher, 111-112  ... 
doi:10.1016/0888-613x(88)90114-4 fatcat:rj373wy2pzff3d3otqdhxs2gca

Microcomputer applications of hybrid intelligent systems

Larry R. Medsker
1996 Journal of Network and Computer Applications  
The integration of neural networks and expert systems has proven to be a useful way to develop real-world applications, and other intelligent technologies including fuzzy logic, genetic algorithms, and  ...  This article presents an analysis of the strengths and limitations of these five intelligent technologies, models for hybrid systems, and a review of the status and significance of research on and development  ...  A less-explored area of integration is the use of the two technologies to produce systems with more general learning abilities such as the semantic interpretation of symbols and the understanding of system  ... 
doi:10.1006/jnca.1996.0015 fatcat:jctv4s7chvbj7jteqbgwittxfu

In honour of Keith Geddes on his 60th birthday

Mark W. Giesbrecht, Stephen M. Watt
2011 Journal of symbolic computation  
Another paper (Carvajal et al., 2005) develops a new hybrid symbolic-numeric method for the fast and accurate evaluation of multiple integrals.  ...  It exploits the increased problem-solving power that can be realized by a hybrid approach combining traditional numerical computation with automated symbolic analysis.  ... 
doi:10.1016/j.jsc.2010.08.009 fatcat:xs5binjdpngqfiricc5yfiaqva
« Previous Showing results 1 — 15 out of 19,983 results