This paper introduces a new wavelet frames-based functional estimation method (i.e. a wavelet-based neural network) which works for more than one dimension functions. The use of frames and wavelets in our approach yields to robust decomposition with an interesting parsimonious property: compression of information in few coefficients. This approach is illustrated using the problem of estimating radioactivity in Chernobyl area.
CITATION STYLE
Soltani, S., Canu, S., Boichu, D., & Grandvalet, Y. (1997). Wavelet frames based estimator. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1327, pp. 320–324). Springer Verlag. https://doi.org/10.1007/bfb0020174
Mendeley helps you to discover research relevant for your work.