[PDF][PDF] Neural networks application to spatial interpolation of climate variables
AP Silva - Carried Out By, STSM on the Framework of COST, 2003 - researchgate.net
Carried Out By, STSM on the Framework of COST, 2003•researchgate.net
As a first approach Artificial Neural Networks (ANN) can be understood as adaptable
systems which can derive relationships between different sets of data from successful cases
of training. But in supervised networks there are many parameters to handle and there is
many times no straightforward procedure given a-priori showing a “black box” behaviour,
although ANN can solve difficult problems that can not mathematically formulated in a
simple way. Often ANN can be applied in regression or classification problems as it is e, g …
systems which can derive relationships between different sets of data from successful cases
of training. But in supervised networks there are many parameters to handle and there is
many times no straightforward procedure given a-priori showing a “black box” behaviour,
although ANN can solve difficult problems that can not mathematically formulated in a
simple way. Often ANN can be applied in regression or classification problems as it is e, g …
As a first approach Artificial Neural Networks (ANN) can be understood as adaptable systems which can derive relationships between different sets of data from successful cases of training. But in supervised networks there are many parameters to handle and there is many times no straightforward procedure given a-priori showing a “black box” behaviour, although ANN can solve difficult problems that can not mathematically formulated in a simple way. Often ANN can be applied in regression or classification problems as it is e, g, presented in this study in reproducing continuous climate fields in space.
researchgate.net
以上显示的是最相近的搜索结果。 查看全部搜索结果