Abstract
There is a gap between two different modes of computation: the symbolic mode and the subsymbolic (neuron-like) mode. The aim of this paper is to overcome this gap by viewing symbolism as a high-level description of the properties of (a class of) neural networks. Combining methods of algebraic semantics and non-monotonic logic, the possibility of integrating both modes of viewing cognition is demonstrated. The main results are (a) that certain activities of connectionist networks can be interpreted as non-monotonic inferences, and (b) that there is a strict correspondence between the coding of knowledge in Hopfield networks and the knowledge representation in weight-annotated Poole systems. These results show the usefulness of non-monotonic logic as a descriptive and analytic tool for analyzing emerging properties of connectionist networks. Assuming an exponential development of the weight function, the present account relates to optimality theory – a general framework that aims to integrate insights from symbolism and connectionism. The paper concludes with some speculations about extending the present ideas.
Similar content being viewed by others
References
A. d’Avila Garcez K. Broda D. Gabbay (2001) ArticleTitle‘Symbolic Knowledge Extraction from Trained Neural Networks: A Sound Approach’ Artificial Intelligence 125 153–205
C. Balkenius P. Gärdenfors (1991) Nonmonotonic Inferences in Neural Networks J.A. Allen R. Fikes E. Sandewall (Eds) Principles of Knowledge Representation and Reasoning Morgan Kaufmann San Mateo, CA
R. Bartsch (2002) Consciousness Emerging John Benjamins Amsterdam & Philadelphia
W. Bechtel (2002) Connectionism and the Mind Blackwell Oxford
P. Boersma B. Hayes (2001) ArticleTitle‘Empirical Tests of the Gradual Learning Algorithm’ Linguistic Inquiry, 32 45–86 Occurrence Handle10.1162/002438901554586
B. Boutsinas M. Vrahatis (2001) ArticleTitle‘Artificial Nonmonotonic Neural Networks’ Artificial Intelligence 132 1–38 Occurrence Handle10.1016/S0004-3702(01)00126-6
N. Chomsky M. Halle (1968) The Sound Pattern of English Harper and Row New York
M. A. Cohen S. Grossberg (1983) ArticleTitleAbsolute Stability of Global Pattern Formation and Parallel Memory Storage by Competetive Neural Networks IEEE Transactions on Systems, Man, and Cybernetics SMC-13 815–826
D. C. Dennett (1995) Darwin’s Dangerous Idea Simon & Schuster New York
M. Derthick (1990) ArticleTitle‘Mundane Reasoning by Settling on a Plausible Model’ Artificial Intelligence 46 107–157 Occurrence Handle10.1016/0004-3702(90)90006-L
J. A. Fodor Z. W. Pylyshyn (1988) ArticleTitle‘Connectionism and Cognitive Architecture: A Critical Analysis’ Cognition 28 3–71 Occurrence Handle10.1016/0010-0277(88)90031-5
R. Frank G. Satta (1998) ArticleTitle‘Optimality Theory and the Generative Complexity of Constraint Violability’ Computational Linguistics 24 307–315
D. Gabbay (1985) ‘Theoretical Foundations for Non-monotonic Reasoning in Expert Systems’ K. Apt (Eds) Logics and Models of Concurrent Systems Springer-Verlag Berlin 439–459
C. Glymour (2001) The Mind’s Arrows The MIT Press Cambridge & London
S. Grossberg (1989) ArticleTitle‘Nonlinear Neural Networks: Principles, Mechanisms, and Architectures’ Neural Networks 1 17–66
S. Grossberg (1996) ArticleTitle‘The Attentive Brain’ American Scientist 83 438–449
J. A. Hendler (1989) ArticleTitle‘Special Issue: Hybrid Systems (Symbolic/Connectionist)’ Connection Science 1 227–342
J. A. Hendler (1991) Developing Hybrid Symbolic/Connectionist Models J. Barnden J. Pollack (Eds) High-level connectionist Models, Advances in Connectionist and Neural Computation Theory, Vol. 1 Ablex Publ. Corp Norwood, NJ
G. E. Hinton T. J. Sejnowski (1983) ‘Optimal Perceptual Inference’ in Proceedings of the Institute of Electronic and Electrical Engineers Computer Society Conference on Computer Vision and Pattern Recognition IEEE Washington, DC 448–453
Hinton, G. E. and T. J. Sejnowski: 1986, ‘Learning and Relearning in Boltzman Machines’, in D. E. Rumelhart, J. L. McClelland, and the PDP research group, pp. 282–317.
J. J. Hopfield (1982) ArticleTitle‘Neural Networks and Physical Systems with Emergent Collective Computational Abilities’ Proceedings of the National Academy of Sciences 79 2554–2558
L. Karttunen (1998) The Proper Treatment of Optimality in Computational Phonology Manuscript. Xerox Research Centre Europe
Kean, M. L.: 1975, The Theory of Markedness in Generative Grammar, Ph.D. thesis, MIT, Cambridge, Mass.
M. L. Kean (1981) ‘On a Theory of Markedness’ R. Bandi A. Belletti L. Rizzi (Eds) Theory of markedness in Generative Grammar Estratto Pisa 559–604
B. Kokinov (1997) ‘Micro-level Hybridization in the Cognitive Architecture DUAL’ R. Sun F. Alexander (Eds) Connectionist-symbolic Integration: From Unified to Hybrid Approaches Lawrence Erlbaum Associates Hilsdale NJ 197–208
S. Kraus D. Lehmann M. Magidor (1990) ArticleTitle‘Nonmonotonic Reasoning, Preferential Models and Cumulative Logics’ Artificial Intelligence 44 167–207 Occurrence Handle10.1016/0004-3702(90)90101-5
W. S. McCulloch W. Pitts (1943) ArticleTitle‘A Logical Calculus of the Ideas Immanent in Nervous Activity’ Bulletin of Mathematical Biophysics 5 115–133
B. Partee H. L. W. Hendriks (1997) Montague Grammar J. Benthem Particlevan A. Meulen Particleter (Eds) Handbook of Logic and Language MIT Press Cambridge 5–91
G. Pinkas (1995) ArticleTitle‘Reasoning, Nonmonotonicity and Learning in Connectionist Networks that Capture Propositional Knowledge’ Artificial Intelligence 77 203–247 Occurrence Handle10.1016/0004-3702(94)00032-V
S. Pinker A. Prince (1988) ArticleTitle‘On Language and Connectionism: Analysis of a Parallel Distributed Processing Model of Language Acquisition’ Cognition 28 73–193
D. Poole (1988) ArticleTitle‘A Logical Framework for Default Reasoning’ Artificial Intelligence, 36 27–47 Occurrence Handle10.1016/0004-3702(88)90077-X
D. Poole (1996) Who Chooses the Assumptions? P. O’Rorke (Eds) Abductive Reasoning MIT Press Cambridge
Prince, A. and P. Smolensky: 1993, Optimality Theory: Constraint Interaction in Generative Grammar. Technical Report CU-CS-696-93, Department of Computer Science, University of Colorado at Boulder, and Technical Report TR-2, Rutgers Center for Cognitive Science, Rutgers University, New Brunswick, NJ.
Rumelhart, D. E., J. L. McClelland, and the PDP research group: 1986, Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Volume I and II. MIT Press/Bradford Books, Cambridge, MA.
D. E. Rumelhart P. Smolensky J. L. McClelland G. E. Hinton (1986) ‘Schemata and Sequential Thought Processes in PDP Models’ D. E. Rumelhart J. L. McClelland (Eds) Parallel Distributed Processing: Explorations in the Microstructure of Cognition. Volume II MIT Press/Bradford Books Cambridge, MA 7–57
L. Shastri V. Ajjanagadde (1993) ArticleTitle‘From Simple Associations to Systematic Reasoning’ Behavioral and Brain Sciences, 16 417–494
L. Shastri C. Wendelken (2000) ‘Seeking Coherent Explanations–A Fusion of Structured Connectionism Temporal Synchrony and Evidential Reasoning Proceedings of Cognitive Science Philadelphia, PA
R. M. Shiffrin W. Schneider (1977) ArticleTitle‘Controlled and Automatic Human Information Processing: II. Perceptual Learning, Automatic Attending, and A General Theory’ Psychological Review 84 127–190 Occurrence Handle10.1037//0033-295X.84.1.1
P. Smolensky (1986) ‘Information Processing in Dynamical Systems: Foundations of Harmony Theory’ D. E. Rumelhart J. L. McClelland (Eds) Parallel distributed processing: Explorations in the microstructure of cognition. Volume I MIT Press/Bradford Books Cambridge, MA 194–281
P. Smolensky (1988) ArticleTitle‘On the Proper Treatment of Connectionism’ Behavioral and Brain Sciences 11 1–23
P. Smolensky (1990) ArticleTitle‘Tensor Product Variable Binding and the Representation of Symbolic Structures in Connectionist Networks’ Artificial Intelligence 46 159–216 Occurrence Handle10.1016/0004-3702(90)90007-M
P. Smolensky (1996) ‘Computational, Dynamical, and Statistical Perspectives on the Processing and Learning Problems in Neural Network Theory’ P. Smolensky M. C. Mozer D. E. Rumelhart (Eds) Mathematical Perspectives on Neural Networks Lawrance Erlbaum Publishers Mahwah, NJ 1–13
P. Smolensky (2000) ArticleTitle‘Grammar-Based Connectionist Approaches to Language’ Cognitive Science 23 589–613
Smolensky, P., and G. Legendre (to appear), The Harmonic Mind: From Neural Computation to optimality-theoretic grammar, Blackwell, Oxford.
Smolensky, P., G. Legendre, and Y. Miyata: 1992, ‘Principles for An Integrated Connectionist/Symbolic Theory of Higher Cognition’, Technical Report CU-CS-600-92, Department of Computer Science, Institute of Cognitive Science, University of Colorado Boulder.
B. Tesar P. Smolensky (2000) Learnability in Optimality Theory MIT Press Cambridge, MA
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Blutner, R. Nonmonotonic Inferences and Neural Networks. Synthese 142, 143–174 (2004). https://doi.org/10.1007/s11229-004-1929-y
Issue Date:
DOI: https://doi.org/10.1007/s11229-004-1929-y