Two approaches were explored which integrate neural net classifiers with Hidden Markov Model (HMM) speech recognizers. Both attempt to improve speech pattern discrimination while retaining the temporal processing advantages of HMMs. One approach used neural nets to provide second-stage discrimination following an HMM recognizer. On a small vocabulary task, Radial Basis Function (RBF) and back-p...

Shao-Han Liu Jzau-Sheng Lin, MEMBER SPIE National Chin-Yi Institute of Technology Department of Electronic Engineering No. 35, Lane 215, Sec. 1, Chung-Shan Rd Taiping, Taichung, Taiwan E-mail: [email protected] Abstract. A new fuzzy Hopfield-model net based on rough-set reasoning is proposed for the classification of multispectral images. The main purpose is to embed a rough-set learning...

By network reliability, we mean the probability of survival of a network for a speciÞed period of time. Obtaining this probability boils down to the formulation and the evaluation of the reliability polynomial. This is a formidable task for networks with a large number of nodes, and an almost impossible task if the node life-times cascade and interact. Consequently, implementing a mathematical ...

1. Introduction The main features of artificial neural networks are the large number of nonlinear processing elements and the massively parallel interconnections among them. Many researchers have studied the hardware required for artificial neural networks and the software for such highly parallel computations. In terms of the hardware, two different approaches have been proposed: VLSI-based ne...

We present a McCulloch-Pitts neural net to recognize even linear languages. The language class is studied in order to define the net topology. Finally, the equivalence between the language class and the languages recognized by the net is proved.

This paper presents a sample of what evolved neural net circuit modules using the socalled ”CoDi-1Bit” neural net model can do. This work is part of an 8 year research project at ATR which aims to build an artificial brain containing a billion neurons by the year 2001, that will be used to control the behaviors of a kitten robot ”Robokoneko”. It looks as though the figure is more likely to be 4...

– gafchromic xrct radiochromic film is a self developing high sensitivity radiochromic filmproduct which can be used for assessment of delivered radiation doses which could match applications suchas computed tomography (ct) dosimetry. the dose response of gafchromic xrct radiochromic film hasbeen measured with reflectance spectrophotometry and desktop scanners. the film automatically changescol...

Retrieved profiles of temperature, water vapor, and cloud liquid water are obtained through neural net inversions of the brightness temperatures, where the neural net is trained using radiosonde soundings and corresponding forward modeled brightness temperatures. An observation operator has also been developed for the direct assimilation of the measured brightness temperatures, using the forwar...

This paper describes the application of neural nets to the management of structural safety at ISMES. A neural net was developed to deal with monitoring data evaluation, in order to perform heuristic interpretation of the data. The net achieves the same results of symbolic processors previously used, but with reduced development and tuning effort.

We investigate the construction, training and application of a neural net for assessing element shape quality of practical unstructured grids arising in mesh generation, adaptive refinement and moving grid applications. Results of numerical experiments are included to validate the process and demonstrate performance of the neural net for both triangulations in 2D and tetrahedral tessellation in...