Answering the connectionist challenge: a symbolic model of learning the past tenses of English verbs

@article{Ling1993AnsweringTC,
  title={Answering the connectionist challenge: a symbolic model of learning the past tenses of English verbs},
  author={Charles X. Ling and Marin Marinov},
  journal={Cognition},
  year={1993},
  volume={49},
  pages={235-290}
}
Learning the Past Tense of English Verbs: The Symbolic Pattern Associator vs. Connectionist Models
  • C. Ling
  • Computer Science
    J. Artif. Intell. Res.
  • 1994
TLDR
This paper presents a general-purpose Symbolic Pattern Associator (SPA) based upon the decision-tree learning algorithm ID3 and concludes that the SPA generalizes the past tense of unseen verbs better than ANN models by a wide margin.
Rules vs. analogy in English past tenses: a computational/experimental study
Learning the Past Tense of English Verbs: Connectionism Fights Back
TLDR
This paper shows that the conclusion that the SPA generalizes the past tense of unseen verbs better than ANN models by a wide margin was based on comparisons with an uncharacteristically poorly performing ANN.
Learning of rules that have high-frequency exceptions: New empirical data and a hybrid connectionist model
  • J. Kruschke, M. A. Erickson
  • Computer Science
    Proceedings of the Sixteenth Annual Conference of the Cognitive Science Society
  • 2019
TLDR
A new connectionist model is presented that combines an exemplar-based module in a connectionist architecture and allowing the system to learn which module should be responsible for which instances, using the competitive gating mechanism introduced by Jacobs, Jordan, Nowlan, and Hinton (1991).
Discussion Connections and symbols : closing the gap
TLDR
An interesting symbolic alternative to current connectionist models of language acquisition and believes that its model outperforms the connectionist model presented by MacWhinney and Leinbach (1991) "by a wide margin".
An Exploration of Connectionist and Symbolic Language Acquisition Techniques
The problem of language acquistion was originally studied by psychologists, but now it is an important problem in Artificial Intelligence as well. Rumelhart and McClelland (from now on referred to as
Modeling the learning of the English past tense with memory-based learning
TLDR
A memory-based learning model that is based on analogy was able to model the learning of the English past tense well compared to previous systems, but a more detailed analysis showed a number of results inconsistent with child language data.
A single-mechanism dual-route model of German verb inflection
TLDR
By focusing on the German participle, it is demonstrated that the performance of the model does not depend on the existence of a dominant ‘default’ class, thus obviating the need for in-built assumptions such as specific processing mechanisms based on grammatical class.
The Declarative/Procedural Model of Lexicon and Grammar
  • M. Ullman
  • Psychology
    Journal of psycholinguistic research
  • 2001
TLDR
It is argued that converging evidence from studies that use a range of psycholinguistic and neurolinguistic approaches with children and adults supports the declarative/procedural model of lexicon and grammar.
...
...

References

SHOWING 1-10 OF 68 REFERENCES
Learning the Past Tense of English Verbs: The Symbolic Pattern Associator vs. Connectionist Models
  • C. Ling
  • Computer Science
    J. Artif. Intell. Res.
  • 1994
TLDR
This paper presents a general-purpose Symbolic Pattern Associator (SPA) based upon the decision-tree learning algorithm ID3 and concludes that the SPA generalizes the past tense of unseen verbs better than ANN models by a wide margin.
On learning the past-tenses of English verbs: implicit rules or parallel distributed processing
TLDR
It is shown how rule-like behavior can emerge from the interactions among a network of units encoding the root form to past tense mapping, and how details of the acquisition process not captured by the rule account emerge.
Why No Mere Mortal Has Ever Flown Out to Center Field
TLDR
The experiments serve as a straightforward demonstration that representations of formal grammatical categories and structures are powerful determinants of linguistic behavior, and are not reducible to semantics, phonology, or prescriptive training.
Connectionism and the problem of systematicity
TLDR
It is argued that Connectionism potentially has an explanation for the acquisition of systematic behaviour, which is an issue not addressed in the Classical paradigm since systematicity is built into a Classical architecture.
On the proper treatment of connectionism
Abstract A set of hypotheses is formulated for a connectionist approach to cognitive modeling. These hypotheses are shown to be incompatible with the hypotheses underlying traditional cognitive
Connectionism and cognitive architecture: A critical analysis
...
...