... variables. The OLS algorithm is implemented by introducing a number of auxiliary orthogonal basis vectors defined in a new transformed space. In our case study, the main objective is not to determine a specific nonlinear model structure but to detect and rank the candidate input variables according...

... network activation value neuron criteria weighting factor procedure matrix transit-time curve input variable Jacobian matrix coefficient multiple input order derivative adjustment output layer jth neuron mth output geological formation activation function Introduction Synthetic...

... Input variables are ranked in a decreasing order according to the impact on a given forecast. If we consider that the risk of the project is too high, we will be able to reduce it by focusing on those variables with higher impact and whose uncertainty could be lowered acquiring more data, e.g...