Empirical Modeling
Empirical models seek to obtain a relationship between factors that influence the process and the resulting outcomes. These models require very little knowledge of the actual process and the model form is generic. Artificial neural networks are an example of a machine learning algorithm where the inputs are related to the outputs through a network of neurons. During training, the neuron parameters are adjusted to generate a relationship between influencing factors and the final outcomes.
ISAT and Neural Networks
In Situ Adaptive Tabulation (ISAT) is an alternative to neural networks that is recieving increased attention for desireable characteristics. ISAT is better suited to large-scale problems, functions with discontinuities, explicit bounds on approximation error and approximation derivatives, is adaptive to new data training without re-optimization, and is simple to tune.
Learn More About ISAT