METHOD FOR CLASSIFIER LEARNING FROM A STREAM OF DATA ON A RESOURCE-CONSTRAINED DEVICE

Methods, apparatuses, systems, and computer program products for artificial intelligence and machine learning for resource constrained devices and systems, including for classifier (130) learning from a stream of data. A classifier (130) may include a neural network comprised of a plurality of layers (140, 150, 160) with each layer comprised of a plurality of neurons. The neural network may include a hidden layer (150) comprised of a plurality of hidden neurons (152A, 152B, 152C, 152D, 152E, ..., 152N). In various embodiments, the size of the hidden layer (150) may be constrained and the training of a hidden layer (150) included removing (420) one or more hidden neurons from the hidden layer (150) of the classifier (130).