Aalbert De Vries

Title: Innovations by Aalbert De Vries

Introduction

Aalbert De Vries is an accomplished inventor based in Lawrence Township, NJ (US). He has made significant contributions to the field of signal processing through his innovative work. His expertise lies in developing methods that enhance the capabilities of neural networks.

Latest Patents

Aalbert De Vries holds a patent for a "Method and system for training a neural network with adaptive weight." This invention presents a signal processing system and method that utilizes a neural network incorporating adaptive weight updating and adaptive pruning for tracking non-stationary signals. The method updates the structural parameters of the neural network in principal component space for every new available input sample. The non-stationary signal is recursively transformed into a matrix of eigenvectors with a corresponding matrix of eigenvalues. The method applies principal component pruning, which involves deleting the eigenmodes corresponding to the smallest saliencies, where a sum of the smallest saliencies is less than a predefined threshold level. This approach reduces the effective number of parameters and generally improves generalization. The output is computed using the remaining eigenmodes, and the weights of the neural network are updated using adaptive filtering techniques.

Career Highlights

Aalbert De Vries is associated with the David Sarnoff Research Center, Inc., where he has contributed to various research initiatives. His work has focused on advancing technologies that leverage neural networks for improved signal processing.

Collaborations

Due to space constraints, the collaborations section will be omitted.

Conclusion

Aalbert De Vries is a notable inventor whose work in neural network training methods has the potential to significantly impact the field of signal processing. His innovative approach demonstrates the importance of adaptive techniques in enhancing the performance of neural networks.