Abstract

A long-standing difficulty for connectionist modeling has been how to represent variable-sized recursive data structures, such as trees and lists, in fixed-width patterns. This paper presents a connectionist architecture which automatically develops compact distributed representations for such compositional structures, as well as efficient accessing mechanisms for them. Patterns which stand for the internal nodes of fixed-valence trees are devised through the recursive use of back-propagation on three-layer autoassociative encoder networks. The resulting representations are novel, in that they combine apparently immiscible aspects of features, pointers, and symbol structures. They form a bridge between the data structures necessary for high-level cognitive tasks and the associative, pattern recognition machinery provided by neural networks. 2 J. B. Pollack 1. Introduction One of the major stumbling blocks in the application of Connectionism to higherlevel cognitive tasks, such as Na...

Citations

1313	Finding structure in time - Elman - 1990
1157	The Fractal Geometry of Nature - MANDELBROT - 1983
976	Multilayer feedforward networks are universal approximators - Hornik, Stinchcombe, et al. - 1989
488	Connectionism and cognitive architecture: a critical analysis - Fodor, Pylyshyn - 1988
413	Parallel networks that learn to pronounce English text - Sejnowski, Rosenberg - 1987
317	On learning the past tense of English verbs - Rumelhart, McClelland - 1986
217	On language and connectionism: Analysis of a parallel distributed processing model of language acquisition - Pinker - 1988
208	Distributed representations - Hinton, McClelland, et al. - 1986
156	Serial Order: A parallel Distributed Processing Approach - Jordan - 1986
154	Learning distributed representations of concepts - Hinton - 1986
140	Mechanisms of sentence processing: assigning roles to constituents - McClelland, Kawamoto - 1986
102	Mapping part-whole hierarchies into connectionist networks - Hinton - 1990
99	Massively Parallel Parsing: A Strongly Interactive Model of Natural Language Interpretation - Waltz, Pollack - 1985
73	Experiments on Learning by BackPropagation - Plaut, Nowlan, et al. - 1986
31	Parsnip: A connectionist network that learns natural language grammar from exposure to natural language sentences - Hanson, Kegl - 1987
29	Symbols among the neurons: Details of a connectionist inference architecture - Touretzky, Hinton - 1985
28	Learning internal representations from gray-scale images: An example of extensional programming - Cottrell, Munro, et al. - 1987
26	Context-free parsing in connectionist networks - Fanty - 1985
25	Case-based problem solving with a Large Knowledge Based of Learned Cases - Lehnert - 1987
23	Cascaded Back Propagation on Dynamic Connectionist Networks - Pollack - 1987
22	BoltzCONS: Reconciling Connectionism with the Recursive Nature of Stacks and Trees - Touretzky - 1986
18	C.: Inductive Information Retrieval using Parallel Distributed Computatio - Mozer - 1984
14	The logic of activation functions - Williams - 1989
13	An interactive activation model of the effect of context in perception, Part I. An account of basic findings - McClelland, Rumelhart - 1981
11	An Introduction to Computing with Neural Networks - Lippman
8	Learning Internal Representations through Error Propagation - Rumelhart, Hinton, et al. - 1986
8	D.: Genetic algorithms - Hinton, Sejnowski, et al. - 1992
5	Four capacity models for coarse-coded symbol memories - Rosenfeld, Touretzky - 1988
4	Rule-based processing in a connectionist system for natural language understanding - Selman - 1985
4	Representing and transforming recursive objects in a neural network, or "Trees do grow on Boltzmann machines - Touretzky - 1986
2	A Parallel Natural Language Processing Architecture with Distributed Control - Berg - 1987
2	A context-free connectionist parser which is not connectionist, but then it is not really context-free either - Charniak, Santos - 1989
1	Natural Language Processing Using Spreading Activation and Lateral - Pollack, Waltz - 1982
1	Forming Global Representations with Back Propagation - Miikkulainen, Dyer - 1988
1	Dimensionality Reduction and Constraint in Later Vision - Saund - 1987