论文信息 - A Connectionist Symbol Manipulator that Discovers the Structure of Context-Free Languages

A Connectionist Symbol Manipulator that Discovers the Structure of Context-Free Languages

We present a neural net architecture that can discover hierarchical and recursive structure in symbol strings. To detect structure at multiple levels, the architecture has the capability of reducing symbols substrings to single symbols, and makes use of an external stack memory. In terms of formal languages, the architecture can learn to parse strings in an LR(O) context-free grammar. Given training sets of positive and negative exemplars, the architecture has been trained to recognize many different grammars. The architecture has only one layer of modifiable weights, allowing for a straightforward interpretation of its behavior.

Michael C. Mozer | Sreerupa Das | M. Mozer | Sreerupa Das

[1] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[2] D. Rumelhart. Learning Internal Representations by Error Propagation, Parallel Distributed Processing , 1986 .

[3] C. Lee Giles,et al. Higher Order Recurrent Networks and Grammatical Inference , 1989, NIPS.

[4] John S. Bridle,et al. Training Stochastic Model Recognition Algorithms as Networks can Lead to Maximum Mutual Information Estimation of Parameters , 1989, NIPS.

[5] Richard Rohwer,et al. The "Moving Targets" Training Algorithm , 1989, NIPS.

[6] Jürgen Schmidhuber,et al. Learning Unambiguous Reduced Sequence Descriptions , 1991, NIPS.

[7] Michael C. Mozer,et al. Induction of Multiscale Temporal Structure , 1991, NIPS.

[8] Colin Giles,et al. Learning Context-free Grammars: Capabilities and Limitations of a Recurrent Neural Network with an External Stack Memory (cid:3) , 1992 .

[9] Andreas Stolcke,et al. Hidden Markov Model} Induction by Bayesian Model Merging , 1992, NIPS.

[10] Mark B. Ring. Learning Sequential Tasks by Incrementally Adding Higher Orders , 1992, NIPS.