Neural Networks: Also known as artificial neural networks (ANNs), are a subcategory of machine learning. Loosely based on how neurons signal each other within the human brain, the neural net consists of multiple (up to millions) processing nodes that are densely interconnected and organized into node layers. Each network is built off an input layer, at least one hidden layer, and an output layer. An individual node might be connected to several nodes in the layer beneath it, from which it receives data, and several nodes in the layer above it, to which it sends data. Just like in the human brain, signals travel from the input layer to the last layer (the output layer), possibly after traversing the layers multiple times.
The “signal” at a connection is a real number, and the output of each neuron is computed by some non-linear function of the sum of its inputs. The connections are called edges. Neurons and edges typically have a weight that adjusts as learning proceeds. The weight increases or decreases the strength of the signal at a connection. Neurons may have a threshold such that a signal is sent only if the aggregate signal crosses that threshold. Typically, neurons are aggregated into layers. Different layers may perform different transformations on their inputs.
A huge volume of information needs to feed to the ANNs in order for them to learn. For example, if your goal is to teach Artificial Neural Network to differentiate between two objects (A and B), the training set (the initial information the ANNs learn on) would need to contain thousands of images tagged as ‘object A’ so the network could start learning. Once it has been trained with enough volume of data, it will try to classify future inputs based on what it thinks it’s seeing (or hearing, depending on the data set) throughout the different units. During the training period, the machine’s output is compared to the description provided by humans. The machine is validated when the human description and what the machine thinks it observes overlap. If it’s incorrect, it goes back through the layers to adjust the mathematical equation.
Use cases: Vehicle control, trajectory prediction, process control, general game playing, pattern recognition, radar systems, face identification, signal classification, object recognition and more, gesture, speech, handwritten and printed text recognition, automated trading systems), machine translation, social network filtering, e-mail spam filtering and more.