ABSTRACT
Artificial neural networks (ANNs) are a family of statistical learning models inspired by biological neural networks (BNN). The common thing about both of them is that they are used to estimate or approximate functions that can depend on a large number of inputs and are generally unknown. ANNs are generally presented as systems of interconnected “neurons” which exchange messages between each other just like central nerves system of animals. The connections have numeric weights known as synaptic weights, that can be tuned based on experience, making neural nets adaptive to inputs and capable of learning. Such system can serve well when used as a replacement for human brain in addressing multivariate complex problem of yarn engineering.
