Learning the Past Tense of English Verbs: Connectionism Fights Back
نویسنده
چکیده
The ability to learn the past tense of English verbs has become a benchmark test for cognitive modelling. In a recent paper, Ling (1994) presented a detailed head-to-head comparison of the generalization abilities of a particular Artificial Neural Network (ANN) model and a general purpose Symbolic Pattern Associator (SPA). The conclusion was that the SPA generalizes the past tense of unseen verbs better than ANN models by a wide margin . In this paper we show that this conclusion was based on comparisons with an uncharacteristically poorly performing ANN. A different ANN model is presented which not only out-performs the existing ANN models by a wide margin but also out-performs the SPA by a significant amount. We provide an explanation of how this happens and suggest several ways in which the model can be improved further. This research was supported by the United Kingdom Joint Councils Initiative in Cognitive Science/HCI, Grant number SPG 9029590.
منابع مشابه
Acquisition of Tense by Persian and English Speaking Children Between 2 to 4 Years Aged
This research is case study which was designed to investigate the acquisition of tense by Persian and English speaking children between 2to4 years aged. Four girls were precisely analyzed in order to figure how the tense of verbs is effective in their speaking that learners need to succeed in their daily lives. The subjects were randomly selected to study and the matter of gender was not consid...
متن کاملA Constructivist Neural Network Learns the Past Tense of English Verbs
A constructivist neural network is presented that models the acquisition of the past tense of English verbs. The network constructs its architecture in response to the learning task, corresponding to neurobiological and psychological evidence. The model outperforms other connectionist and symbolic models in learning and in displaying psychologically realistic learning and generalization behavio...
متن کاملLearning the past tense of English verbs using inductive logic programming
This paper presents results on using a new inductive logic programming method called Foidl to learn the past tense of English verbs. The past tense task has been widely studied in the context of the symbolic/connectionist debate. Previous papers have presented results using various neural-network and decision-tree learning methods. We have developed a technique for learning a special type of Pr...
متن کاملInduction of First-Order Decision Lists: Results on Learning the Past Tense of English Verbs
This paper presents a method for inducing logic programs from examples that learns a new class of concepts called rst-order decision lists, de ned as ordered lists of clauses each ending in a cut. The method, called Foidl, is based on Foil (Quinlan, 1990) but employs intensional background knowledge and avoids the need for explicit negative examples. It is particularly useful for problems that ...
متن کاملPast Tenses of Verbs and First-order Learning
Learning to transform English verbs from present to past tense has been studied extensively in the connectionist literature. A recent paper describes a symbolic approach that outperforms neural networks on this task, but the new system is still constrained by propositional-level attribute-value representations. A rst-order learning method that uses a relatively natural representation is found t...
متن کامل