Neural Networks, Data Clustering etc

Citations refer to Papers (since 1989)
My interests in the area of Neural Networks have faded. In the past we developed an efficient learning algorithm for binary-valued neurons [9] , [13] , [T2] , [T4] , proposed and implemented neural networks with purely optical elements [10] , [C2] , [T1] , [T4] and studied the computational capabilities of various single [7] , [8] [7,8] and multilayer [18] , [22] , [T8] perceptrons. Subsequently I shifted more and more towards applications of neural networks to "real" problems, such as color conversion, sensor fusion, classification of (noisy) data [20] , [T7] and time series prediction [T10] . More recently, we applied perceptron learning schemes to the problem area of protein structure prediction.

We discovered a very efficient way to perform non- parametric clustering of data , by using a physical system to provide a robust solution to this problem [28] , [32] , [C4] . This work triggered my entry into my current field of interest - Bioinformatics; in particular - analysis of gene expression data.



Eytan Domany