Ray J. Frank

Time Series Prediction and Neural Networks

Neural Network approaches to time series prediction are briefly discussed, and the need to find the appropriate sample rate and an appropriately sized input window identified. Relevant theoretical results from dynamic systems theory are briefly introduced, and heuristics for finding the appropriate sampling rate and embedding dimension, and thence window size, are discussed. The method is applied to several time series and the resulting generalisation performance of the trained feed-forward neural network predictors is analysed. It is shown that the heuristics can provide useful information in defining the appropriate network architecture.

VISUAL CROWDING AND CATEGORY SPECIFIC DEFICITS FOR PICTORIAL STIMULI : A NEURAL NETWORK MODEL

This paper describes a series of modular neural network simulations of visual object processing. In a departure from much previous work in this domain, the model described here comprises both supervised and unsupervised modules and processes real pictorial representations of items from different object categories. The unsupervised module carries out bottom-up encoding of visual stimuli, thereby developing a “perceptual” representation of each presented picture. The supervised component then classifies each perceptual representation according to a target semantic category. Model performance was assessed (1) during learning, (2) under generalisation to novel instances, and (3) after lesion damage at different stages of processing. Strong category effects were observed throughout the different experiments, with living things and musical instruments eliciting greater recognition failures relative to other categories. This pattern derives from within-category similarity effects at the level of perceptual representation and our data support the view that visual crowding can be a potentially important factor in the emergence of some category-specific impairments. The data also accord with the cascade model of object recognition, since increased competition between perceptual representations resulted in category-specific impairments even when the locus of damage was within the semantic component of the model. Some strengths and limitations of this modelling approach are discussed and the results are evaluated against some other accounts of category-specific recognition failure.

Dimensionality reduction of face images for gender classification

Data in most real world applications are high dimensional and learning algorithms like neural networks have problems in handling high dimensional data. However, the intrinsic dimension is often much less than the original dimension of the data. We use a fractal based method to estimate the intrinsic dimension and show that a nonlinear projection method called curvilinear component analysis can effectively reduce the original dimension to the intrinsic dimension. We apply this approach for dimensionality reduction of the face images data and use neural network classifiers for gender classification.

Analysis of linear and nonlinear dimensionality reduction methods for gender classification of face images

Data in many real world applications are high-dimensional and learning algorithms like neural networks may have problems in handling high-dimensional data. However, the ‘intrinsic dimension (ID)’ is often much less than the original dimension of the data. Here, we use fractal based methods to estimate the ID and show that a nonlinear projection method called curvilinear component analysis (CCA) can effectively reduce the original dimension to the ID. We apply this approach for dimensionality reduction of the face images data and use neural network classifiers for gender classification.

The Role of Global and Feature Based Information in Gender Classification of Faces: A Comparison of Human Performance and Computational Models

Most computational models for gender classification use global information (the full face image) giving equal weight to the whole face area irrespective of the importance of the internal features. Here, we use a global and feature based representation of face images that includes both global and featural information. We use dimensionality reduction techniques and a support vector machine classifier and show that this method performs better than either global or feature based representations alone. We also present results of human subjects performance on gender classification task and evaluate how the different dimensionality reduction techniques compare with human subjects performance. The results support the psychological plausibility of the global and feature based representation.

Input window size and neural network predictors

Neural network approaches to time series prediction are briefly discussed, and the need to specify an appropriately sized input window identified. Relevant theoretical results from dynamic systems theory are briefly introduced, and heuristics for finding the correct embedding dimension, and hence window size, are discussed. The method is applied to two time series and the resulting generalisation performance of the trained feedforward neural network predictors is analysed. It is shown that the heuristics can provide useful information in defining the appropriate network architecture.

Gender Classification of Face Images: The Role of Global and Feature-Based Information

Most computational models of gender classification use global information (the full face image) giving equal weight to the whole face area irrespective of the importance of the internal features. Here we use a two-way representation of face images that includes both global and featural information. We use dimensionality reduction techniques and a support vector machine classifier and show that this method performs better than either global or feature based representations alone.

Using single layer networks for discrete, sequential data: an example from natural language processing

Natural Language Processing (NLP) is concerned with processing ordinary, unrestricted text. This work takes a new approach to a traditional NLP task, using neural computing methods. A parser which has been successfully implemented is described. It is a hybrid system, in which neural processors operate within a rule based framework. The neural processing components belong to the class of Generalized Single Layer Networks (GSLN). In general, supervised, feed-forward networks need more than one layer to process data. However, in some cases data can be pre-processed with a non-linear transformation, and then presented in a linearly separable form for subsequent processing by a single layer net. Such networks offer advantages of functional transparency and operational speed. For our parser, the initial stage of processing maps linguistic data onto a higher order representation, which can then be analysed by a single layer network. This transformation is supported by information theoretic analysis. Three different algorithms for the neural component were investigated. Single layer nets can be trained by finding weight adjustments based on (a) factors proportional to the input, as in the Perceptron, (b) factors proportional to the existing weights, and (c) an error minimization method. In our experiments generalization ability varies little; method (b) is used for a prototype parser. This is available via telnet.

Categorizing facial expressions: a comparison of computational models

Recognizing expressions is a key part of human social interaction, and processing of facial expression information is largely automatic for humans, but it is a non-trivial task for a computational system. The purpose of this work is to develop computational models capable of differentiating between a range of human facial expressions. Raw face images are examples of high-dimensional data, so here we use two dimensionality reduction techniques: principal component analysis and curvilinear component analysis. We also preprocess the images with a bank of Gabor filters, so that important features in the face images may be identified. Subsequently, the faces are classified using a support vector machine. We show that it is possible to differentiate faces with a prototypical expression from the neutral expression. Moreover, we can achieve this with data that has been massively reduced in size: in the best case the original images are reduced to just 5 components. We also investigate the effect size on face images, a concept which has not been reported previously on faces. This enables us to identify those areas of the face that are involved in the production of a facial expression.

Discriminating Angry, Happy and Neutral Facial Expression: A Comparison of Computational Models

Recognizing expressions are a key part of human social interaction, and processing of facial expression information is largely automatic for humans, but it is a non-trivial task for a computational system. The purpose of this work is to develop computational models capable of differentiating between a range of human facial expressions. Raw face images are examples of high dimensional data, so here we use two dimensionality reduction techniques: Principal Component Analysis and Curvilinear Component Analysis. We also preprocess the images with a bank of Gabor filters, so that important features in the face images are identified. Subsequently the faces are classified using a Support Vector Machine. We show that it is possible to differentiate faces with a neutral expression from those with a happy expression and neutral expression from those of angry expressions and neutral expression with better accuracy. Moreover we can achieve this with data that has been massively reduced in size: in the best case the original images are reduced to just 5 components with happy faces and 5 components with angry faces.