The OCaml Journal just published an article about artificial intelligence:
"Artificial neural networks are a relatively simple and early form of artificial intelligence modelled directly on the structure of an organic brain. A neural network is composed of many interconnected neurons. Each neuron acts as a simple linear classifier by summing its weighted inputs and firing an output accordingly. The functions involved are essentially ad-hoc but are usually modelled on those of real neurons. This article describes the design and implementation of a neural network that uses an unconventional functional approach to handle vector-matrix operations that allows temporary data structures to be avoided..."
To read this article and more, subscribe to The OCaml Journal today!