摘要:

Backpropagation learning (BP) is known for its serious limitations in generalizing knowledge from certain types of learning material. In this paper, we describe a new learning algorithm, BP-SOM, which overcomes some of these limitations as is shown by its application to four benchmark tasks. BP-SOM is a combination of a multi-layered feedforward network (MFN) trained with BP and Kohonen's self-organizing maps (SOMs). During the learning process, hidden-unit activations of the MFN are presented as learning vectors to SOMs trained in parallel. The SOM information is used when updating the connection weights of the MFN in addition to standard error backpropagation. The effect of the augmented error signal is that, during learning, clusters of hiddenunit activation patterns of instances associated with the same class tend to become highly similar. In a number of experiments, BP-SOM is shown (i) to improve generalization performance (i.e. avoid overfitting); (ii) to increase the amount of hidden units that can be pruned without loss of generalization performance and (iii) to provide a means for automatic rule extraction from trained networks. The results are compared with results achieved by two other learning algorithms for MFNs: conventional BP and BP augmented with weight decay. From the experiments and the comparisons, we conclude that the hybrid BP-SOM architecture, in which supervised and unsupervised and learning co-operate in finding adequate hidden-layer representations, successfully combines the advantages of supervised and unsupervised learning.

ISSN:0954-0091    

DOI:10.1080/095400997116621

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

It has been claimed that rotation of neuronal population vectors in the primary motor cortex during tasks involving transformation of movement directions provides direct evidence that a mental rotation algorithm is being used. An alternative explanation offered here, the 'asynchronous superposition hypothesis', asserts that the apparent rotation can result from the summation of two stationary vectors, one in the stimulus direction and one in the target movement direction, which vary in length over time. Computer simulations demonstrate the surprising result that the asynchronous superposition hypothesis can produce activation of motor corex cells with intermediate preferred directions, something previously assumed to confirm a mental rotation algorithm. Simulations also demonstrate that the asynchronous superposition hypothesis accounts for some aspects of the data more naturally than does an explanation based on a mental rotation algorithm, and suggest experimental tests that can distinguish between these two hypotheses.

ISSN:0954-0091    

DOI:10.1080/095400997116630

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

In this paper, syntactic parsing is discussed in the context of connectionism, a new model, the confluent preorder parser (CPP), is proposed which exemplifies the holistic parsing paradigm. Holistic parsing has the advantage that little knowledge has to be assumed concerning the detailed parsing algorithm. This algorithm is often unkown or debatable, especially when human language understanding is concerned. In the CPP, syntactic parsing is achieved by transforming from the connectionist representation of the sentence to the connectionist representation of the preorder traversal of its parse tree, instead of to the representation of the parse tree itself. As revealed by the simulation experiments, generalization performance is excellent (as high as 90%). Also, the CPP is capable of parsing erroneous sentences and resolving lexical category ambiguities. A systematic study is conducted to explore the range of factors which can affect the effectiveness of the system. The error-recovery capability is especially useful in natural language processing when incomplete or even ungrammatical sentences must be dealt with.

ISSN:0954-0091    

DOI:10.1080/095400997116649

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor

摘要:

Connectionist modelling is a relatively new and valuable technique for the development of language processing theories. Most of language production models thus far have been discussed with respect to the English language. In this study, we present an interactive activation model of noun production for a morphologically complex language, viz. Finnish. Our model follows the psycholinguistic assumptions of the SAID model, an experimentally corroborated production model. The present model is a local connectionist model and it has been evaluated by computer simulation. It has been tested with respect to case and possessive suffixes of nouns. Furthermore, to test the psycholinguistic plausibility of the model, simulations of the most robust phonological frame constraint in speech errors, the initialness effect, were performed. These simulations showed good correspondence to empirical error data. We argue that the present model provides a basis for further computational research in morphologically rich languages like Finnish.

ISSN:0954-0091    

DOI:10.1080/095400997116658

出版商:Taylor & Francis Group    

年代:1997

数据来源: Taylor