Extracting Reduced Logic Programs from Artificial Neural Networks

Extracting Reduced Logic Programs from Artificial Neural Networks (2005)

Cached

Download Links

by Jens Lehmann , Sebastian Bader , Pascal Hitzler

Venue:	Proceedings of the IJCAI-05 Workshop on Neural-Symbolic Learning and Reasoning, NeSy’05
Citations:	2 - 1 self

BibTeX

@INPROCEEDINGS{Lehmann05extractingreduced,
    author = {Jens Lehmann and Sebastian Bader and Pascal Hitzler},
    title = {Extracting Reduced Logic Programs from Artificial Neural Networks},
    booktitle = {Proceedings of the IJCAI-05 Workshop on Neural-Symbolic Learning and Reasoning, NeSy’05},
    year = {2005}
}

Bookmark

OpenURL

Abstract

Artificial neural networks can be trained to perform excellently in many application areas. While they can learn from raw data to solve sophisticated recognition and analysis problems, the acquired knowledge remains hidden within the network architecture and is not readily accessible for analysis or further use: Trained networks are black boxes. Recent research efforts therefore investigate the possibility to extract symbolic knowledge from trained networks, in order to analyze, validate, and reuse the structural insights gained implicitly during the training process. In this paper, we will study how knowledge in form of propositional logic programs can be obtained in such a way that the programs are as simple as possible --- where simple is being understood in some clearly defined and meaningful way.

Citations

3909	Neural Networks for Pattern Recognition - Bishop - 1995
1739	Foundations of Logic Programming - Lloyd - 1984
544	Fractals Everywhere - Barnsley - 1988
517	UCI repository of machine learning databases - Hettich, Merz - 1998
408	Raedt. Inductive logic programming: Theory and methods - Muggleton, de - 1994
204	Survey and critique of techniques for extracting rules from trained artificial neural networks - Andrews, Diederich, et al. - 1995
164	The MONK's problem: A performance comparison of different learning algorithms - Thrun - 1991
162	Fibring logics - Gabbay - 1999
133	Knowledge-based artificial neural networks - Towell, Shavlik - 1994
48	Approximating the semantics of logic programs by recurrent neural networks - Hölldobler, Kalinke, et al. - 1999
38	Symbolic knowledge extraction from trained neural networks: A sound approach - Garcez, d‟Avila, et al.
37	Logic programs and connectionist networks - Hitzler, Hölldobler, et al.
21	Towards a new massively parallel computational model for logic programming - Hölldobler, Kalinke - 1994
17	Dimensions of neural-symbolic integration - a structured survey - Bader, Hitzler - 2005
16	The connectionist inductive learning and logic programming system - Garcez, Zaverucha - 1999
15	Rule Extraction from Recurrent Neural Networks: A Taxonomy and Review - Jacobsson - 2005
14	Argumentation Neural Networks - Garcez, Gabbay, et al. - 2004
13	Logic programs, iterated function systems, and recurrent radial basis function networks - Bader, Hitzler
13	On approximation in the integration of connectionist and logic-based systems - Seda, Lane - 2005
10	A connectionist inductive learning system for modal logic programming - Garcez, Lamb, et al. - 2002
9	The integration of connectionism and first-order knowledge representations and reasoning as a challenge for artificial inteligence - Bader, Hitzler, et al.
9	A fully connectionist model generator for covered first-order logic programs - Bader, Hitzler, et al.
8	Integrating first-order logic programs and connectionist systems — a constructive approach - Bader, Hitzler, et al. - 2005
8	On approximation of the semantic operators determined by bilattice-based logic programs - Komendantskaya, Seda, et al. - 2005
7	Learning models of predicate logical theories with neural networks based on topos theory - Gust, Kühnberger, et al.
7	de Carvalho. Logical inference and inductive learning in artificial neural networks - Garcez, Zaverucha, et al. - 1997
6	Computing first-order logic programs by fibring artificial neural networks - Bader, Garcez, et al. - 2005
5	Lessons from past, current issues, and future research directions in extracting the knowledge embedded in artificial neural networks. Hybrid Neural Systems - Tickle, Maire, et al. - 1998
4	The core method: Connectionist model generation for first-order logic programs - Bader, Hitzler, et al. - 2006
4	Metalevel priorities and neural networks - Garcez, Broda, et al. - 2000
4	Neural-symbolic intuitionistic reasoning - Garcez, Lamb, et al. - 2003
3	Rule extraction from local cluster neural nets - Andrews, Geva
3	Neural networks and structured knowledge: Rule extraction and applications - Kurfess
3	Corollaries on the fixpoint completion: studying the stable semantics by means of the Clark completion - Hitzler
3	Ontology leaning as a use case for neural-symbolic integration - Hitzler, Bader, et al. - 2005
2	d’Avila Garcez. Computing first-order logic programs by fibring artificial neural network - Bader, Hitzler, et al. - 2005
2	Extracting logic programs from artificial neural networks - Lehmann - 2005
2	Logical inference and inductive learning in artificial neural networks - Garcez, Zaverucha, et al. - 1997
2	Computation of normal logic programs by fibring neural networks - Komendantsky, Seda - 2005
1	Ontology learning as a use-case for neural-symbolic intergration - Hitzler, Bader, et al. - 2005
1	Knowledge extraction using artificial neural networks: application to radar target identification - Remm, Alexandre
1	Fibring Logics. Oxford Univesity Press - Gabbay - 1999