We compare two classifier approaches, namely classifiers based on Multi Layer Perceptrons (MLPs) and Gaussian Mixture Models (GMMs), for use in a face verification system. The comparison is carried out in terms of performance, robustness and practicability. Apart from structural differences, the two approaches use different training criteria; the MLP approach uses a discriminative criterion, wh...

We introduce the notion of monotone linear-programming circuits (MLP circuits), a model of computation for partial Boolean functions. Using this model, we prove the following results. 1. MLP circuits are superpolynomially stronger than monotone Boolean circuits. 2. MLP circuits are exponentially stronger than monotone span programs. 3. MLP circuits can be used to provide monotone feasibility in...

One possible definition of the length of a digitized curve in 3D is the length of the shortest polygonal curve lying entirely in a cube curve. In earlier work the authors proposed an iterative algorithm for the calculation of this minimal length polygonal curve (MLP). This paper reviews the algorithm and suggests methods to speed it up by reducing the set of possible locations of vertices of th...

The adenovirus L4-22K protein both activates and suppresses transcription from the adenovirus major late promoter (MLP) by binding to DNA elements located downstream of the MLP transcriptional start site: the so-called DE element (positive) and the R1 region (negative). Here we show that L4-22K preferentially binds to the RNA form of the R1 region, both to the double-stranded RNA and the single...

Feature projection by non-linear discriminant analysis (NLDA) can substantially increase classification performance. In automatic speech recognition (ASR) the projection provided by the pre-squashed outputs from a one hidden layer multi-layer perceptron (MLP) trained to recognise speech subunits (phonemes) has previously been shown to significantly increase ASR performance. An analogous approac...

Multi-layer feedforward neural networks with sigmoidal activation functions have been termed "universal function approximators". Although these types of networks can approximate any continuous function to a desired degree of accuracy, this approximation may require an inordinate number of hidden nodes and is only accurate over a finite interval. These short comings are due to the standard multi...

This paper proposes a new investigation on Gaussian mixture model (GMM) by comparing it with some preliminary experiments on multilayered perceptron network (MLP) with backpropagation learning algorithm (BKP) and dynamic time warping (DTW) techniques on Thai text-dependent speaker identification system. Three major identification engines are conducted on 50 speakers with isolated digits 0-9. Tr...

The performance of global affine and nonlinear transformations for speaker adaptation in a hidden Markov model (HMM) speech recognition system are compared in this paper. The nonlinear transformation was obtained with a multilayer perceptron network (MLP) which was trained during the adaptation process to transform the mean vectors of the HMMs such that the output probabilities of the HMMs for ...