Fourth International Workshop on

Neural-Symbolic Learning and Reasoning

July 21st, 2008

NeSy'05 took place
at IJCAI-05, Edinburgh,
Scotland, 1st of August 2005.

NeSy'06 took place
at ECAI2006, Riva del Garda,
Italy, 29th of August 2006.

NeSy'07 took place
at IJCAI-07, Hydarabad, India, 8th of January 2007.

09.00 - 10.00 | Keynote by Kai-Uwe Kühnberger: Modeling Reasoning Mechanisms by Neural-Symbolic Learning. |

coffee break | |

10:30 - 10:50 | Ekaterina Komendantskaya: Unification by Error-Correction |

10:55 - 11:10 | Matthew Cook: The Reusable Symbol Problem (Position Paper) |

11:15 - 11:35 | Claudine Brucks, Michael Hilker, Christoph Schommer, Cynthia Wagner, Ralph Weires: Symbolic Computing with Incremental Mind-maps to Manage and Mine Data Streams - Some Applications |

11:40 - 12:00 | Sebastian Bader, Steffen Hölldobler, Nuno C. Marques: Guiding Backprop by Inserting Rules |

12:05 - 12:25 | Tsvi Achler, Eyal Amir: Hybrid Classification and Symbolic-Like Manipulation Using Self-Regulatory Feedback Networks |

**Kai-Uwe Kühnberger, Osnabrück: Modeling Reasoning Mechanisms by Neural-Symbolic Learning**

Currently, neural-symbolic integration covers - at least in theory - a whole bunch
of types of reasoning: neural representations (and partially also neural-inspired
learning approaches) exist for modeling propositional logic (programs), whole
classes of manyvalued logics, modal logic, temporal logic, and epistemic logic,
just to mention some important examples [2,4]. Besides these propositional
variants of logical theories, also first proposals exist for approximating *infinity*
with neural means, in particular, theories of first-order logic. An example is the
core method intended to learn the semantics of the single-step operator TP for
first-order logic (programs) with a neural network [1]. Another example is the
neural approximation of variable-free first-order logic by learning representations
of arrow constructions (which represent logical expressions) in the Rn using
Topos constructions [3].

Although these examples show a certain success of neural-symbolic learning and reasoning research, there are several non-trivial challenges. First, there exist a variety of frameworks that seem to have rather different and seemingly incompatible foundations. Second, potential application domains where the strengths of neural-symbolic integration could be documented and its potential be shown are not really known. Third, the conceptual understanding of the cognitive relevance and the cognitive plausibility of neural-symbolic learning and reasoning should be clarified. In this talk, I will address these questions and propose some ideas for answers. I will sketch general assumptions of solutions for the neuralsymbolic modeling of a variety of logical reasoning mechanisms. Then I will propose some application domains where neural-symbolic frameworks can be successfully applied. I will finish the talk with some speculations concerning cognitively relevant implications and the degree of cognitive plausibility of neuralsymbolic learning and reasoning in general.

**References**

[1] S. Bader, P. Hitzler, S. Hölldobler and A. Witzel: A Fully Connectionist Model
Generator for Covered First-Order Logic Programs. In Proceedings of the
Twentieth International Joint Conference on Artificial Intelligence, 2007, pp. 666-671.

[2] A. d'Avila Garcez, L. Lamb and D. Gabbay: Neural-Symbolic Cognitive
Reasoning, Cognitive Technologies, Springer, 2008.

[3] H. Gust, K.-U. Kühnberger and P. Geibel: Learning Models of Predicate Logical
Theories with Neural Networks based on Topos Theory. In P. Hitzler and B.
Hammer (eds.): Perspectives of Neural-Symbolic Integration, Studies in
Computational Intelligence (SCI) 77, Springer, 2007, pp. 233-264.

[4] E. Komendantskaya, M. Lane and A. Seda: Conenctionist Representation of
Multi-Valued Logic Programs. In P. Hitzler and B. Hammer (eds.): Perspectives
of Neuro-Symbolic Integration, Studies in Computational Intelligence (SCI) 77,
Springer, 2007, pp. 283-313.

Artificial Intelligence researchers continue to face huge challenges in their quest to develop truly intelligent systems. The recent developments in the field of neural-symbolic integration bring an opportunity to integrate well-founded symbolic artificial intelligence with robust neural computing machinery to help tackle some of these challenges.

The Workshop on Neural-Symbolic Learning and Reasoning is intended to create an atmosphere of exchange of ideas, providing a forum for the presentation and discussion of the key topics related to neural-symbolic integration. Topics of interest include:

- The representation of symbolic knowledge by connectionist systems;
- Learning in neural-symbolic systems;
- Extraction of symbolic knowledge from trained neural networks;
- Reasoning in neural-symbolic systems;
- Biological inspiration for neural-symbolic integration;
- Neural networks and probabilities;
- Applications in robotics, semantic web, engineering, bioinformatics, etc.

Researchers and practitioners are invited to submit original papers that have not been submitted for review or published elsewhere. Submitted papers must be written in English and should not exceed 5 pages in the case of research and experience papers, and 2 pages in the case of position papers (including figures, bibliography and appendices) in ECAI2008 format as described in the ECAI2008 submissions and style guide. All submitted papers will be judged based on their quality, relevance, originality, significance, and soundness. Papers must be submitted directly by email in PDF format to nesy@soi.city.ac.uk

Selected papers will have to be presented during the workshop. The workshop will include extra time for audience discussion of the presentation allowing the group to have a better understanding of the issues, challenges, and ideas being presented.

Accepted papers will be published electronically in the CEUR workshop proceedings (bearing an ISSN number). Hardcopies will be distributed during the workshop. Authors of the best papers will be invited to submit a revised and extended version of their papers to the Journal of Logic and Computation, OUP.

Deadline for submission: May 16th, 2008

Notification of acceptance: May 30th, 2008

Camera-ready paper due: June 6th, 2008

Workshop date: July 21st, 2008

ECAI2008 main conference dates: 21st to 25th of July, 2008

Artur d'Avila Garcez (City University London, UK)

Pascal Hitzler (University Karlsruhe, Germany)

Sebastian Bader, TU Dresden, Germany

Howard Blair, Syracuse University, U.S.A.

Luc de Raedt, KU Leuven, Belgium

Marco Gori, University of Siena, Italy

Barbara Hammer, TU Clausthal, Germany

Ioannis Hatzilygeroudis, University of Patras, Greece

Steffen Hölldobler, TU Dresden, Germany

Ekaterina Komendantskaya, Sophia Antipolis, France

Kai-Uwe Kühnberger, Osnabrück, Germany

Luis Lamb, Federal University of Rio Grande do Sul, Brazil

Roberto Prevete, University of Naples, Italy

Dan Roth, University of Illinois at Urbana-Champaign, U.S.A.

Anthony K. Seda, University College Cork, Ireland

Frank van der Velde, Leiden University, The Netherlands

Gerson Zaverucha, Federal University of Rio de Janeiro, Brazil

Kai-Uwe Kühnberger, Osnabrück

General questions concerning the workshop should be addressed to nesy@soi.city.ac.uk.