Autonomously acting cars and predicting market behaviour: some application scenarios for ANNs

Abstract

As already discussed, deep artificial neural networks are, in some aspects, the most humanlike artificially developed way of processing information we have thitherto. Very impressive and important achievements have been made due to complex network structures and sophisticated training algorithms, and what can already be achieved is not so far from the capabilities of what parts of the human brain, such as the visual cortex, can achieve. However until recently, the typical research in that area concerned problems that are perfectly suitable for being processed on a computer, like prediction of numerical values or clustering of unstructured data, and individual approaches are simply not potent enough for approaching the full stack of a human brain’s capabilities.