An adaptive learning architecture for modeling manufacturing processes involving several control variables is described. The use of this architecture to process modeling and recipe synthesis for deposition rate, stress, and film thickness in low-pressure chemical vapor deposition (LPCVD) of undoped polysilicon is discussed. In this architecture the model for a process is generated by combining the qualitative knowledge of human experts, captured in the form of influence diagrams, and the learning abilities of neural networks for extracting the quantitative knowledge that relates the parameters of a process. To evaluate the merits of this methodology, the accuracy of these new models is compared to that of more conventional models generated by the use of first principles and/or statistical regression analysis. The models generated by the integration of influence diagrams and neural networks are shown to have half the error or less, even though given only half as much information in creating the models. Furthermore, it is shown that, by employing the generalization ability of neural networks in the synthesis algorithm, new recipes can be produced for the process. Two such recipes are generated for the LPCVD process. One is a zero-stress polysilicon film recipe; the second is a uniform deposition rate recipe which is based on the use of a nonuniform temperature distribution during deposition. >

Use of influence diagrams and neural networks in modeling semiconductor manufacturing processes

A recurrent neural network controller and learning algorithm for the on-line learning control of autonomous underwater vehicles

The ALOPEX process is a broadly defined optimization procedure which simultaneously varies several parameters based on single value feedback and noise application, ALOPEX is used to solve an image recognition problem using several hexagonally based templates. Efficiency improvements are examined, such as a 'temperature' effect, arrival at optimal parameter values, and a form of parallelization. Two different forms of the 'cost' function are analyzed, and a hierarchical system is attempted to attain faster 'good' solutions. The analysis is expanded to multiple template simulations, where the algorithm is able to make appropriate recognition decisions on noisy or incomplete data. >

A pattern recognition application of the ALOPEX process with hexagonal arrays

In this paper we view learning as an unconstrained non-linear minimization problem in which the objective function is defined by the negative log-likelihood function and the search space by the parameter space of an origin constrained product unit neural spatial interaction model. We consider Alopex based global search, as opposed to local search based upon backpropagation of gradient descents, each in combination with the bootstrapping pairs approach to solve the maximum likelihood learning problem. Interregional telecommunication traffic flow data from Austria are used as test bed for comparing the performance of the two learning procedures. The study illustrates the superiority of Alopex based global search, measured in terms of Kullback and Leibler's information criterion.

/pdf/learning-in-neural-spatial-interaction-models-a-statistical-44up28t0ud.pdf

Learning in neural spatial interaction models: A statistical perspective

In this paper a novel modular product unit neural network architecture is presented to model singly constrained spatial interaction flows. The efficacy of the model approach is demonstrated for the origin constrained case of spatial interaction using Austrian interregional telecommunication traffic data. The model requires a global search procedure for parameter estimation, such as the Alopex procedure. A benchmark comparison against the standard origin constrained gravity model and the two–stage neural network approach, suggested by Openshaw (1998), illustrates the superiority of the proposed model in terms of the generalization performance measured by ARV and SRMSE.

Neural Network Modeling of Constrained Spatial Interaction Flows: Design, Estimation, and Performance Issues

The Alopex process has been applied to a variety of problems in which maxima or minima of functions of many parameters are to be determined. We present here different forms of the algorithm and a simple theory of the dynamics of the process. Computer simulations of several applications are presented to illustrate the versatility of the method. An artificial visual system is described in which partial stimuli are completed and patterns are created out of noise with only a scalar cost function to guide the process.

E. Harth

Papers

Dynamics of Alopex Process: Application to Optimization Problems