Supervised relational learning over labeled graphs, e.g. via recursive neural nets, received considerable attention from the connectionist community. Surprisingly, with the exception of recursive self organizing maps, unsupervised paradigms have been far less investigated. In particular, no algorithms for density estimation over graphs are found in the literature. This paper introduces first a ...

The classical non-recursive methods to estimate unknown parameters of the model, such as the maximum likelihood method, the method of least squares etc. eventually require maximization procedures. These methods are often difficult to implement, especially for non i.i.d. models. If for every sample size n, when new data are acquired, an estimator has to be computed afresh, and if a numerical met...

This paper is concerned with the asymptotic behaviour of estimation procedures which are recursive in the sense that each successive estimator is obtained from the previous one by a simple adjustment. The results of the paper can be used to determine the form of the recursive procedure which is expected to have the same asymptotic properties as the corresponding nonrecursive one defined as a so...

This paper presents a new pel recursive motion compensated prediction algorithm for video coding applications. The derivation of the algorithm is based on Recursive Least Squares (RLS) estimation that minimizes the mean square prediction error for each pel (picture element). A comparison with the modified Steepest-descent gradient estimation algorithm shows significant improvement in terms of m...