The evaluation of a standard Gaussian process regression model takes time linear in the number of training data points. In this paper, the models are approximated in the feature sp...
We show that many rational parametric curves can be interpolated, in a Hermite sense, by polynomial curves whose degree, relative to the number of data being interpolated, is lowe...
We study the problem of minimizing the expected loss of a linear predictor while constraining its sparsity, i.e., bounding the number of features used by the predictor. While the r...
The class of variable-length finite-state joint source-channel codes is defined and a polynomial complexity algorithm for the evaluation of their distance spectrum presented. Issu...