Metric Learning - Metric learning learns a distance or similarity representation from data so that comparisons better reflect a specified task.
Modern Hopfield network - A modern Hopfield network is an associative-memory model that extends classical Hopfield ideas through different energy functions and retrieval dynamics.
Multi-surface method - The multi-surface method is a historical approach to classifying patterns by constructing several separating surfaces.
Multidimensional Scaling - Multidimensional scaling, or MDS, is a family of methods that represents objects as points so that distances between the points reflect supplied...
Multilayer perceptron - A multilayer perceptron, or MLP, is a feedforward neural network built from layers of connected computational units.
Multimodal Deep Learning - Multimodal deep learning uses deep neural networks to learn from relationships among different kinds of data, such as text, images, audio or sensor signals.
Multimodal Learning - Multimodal learning is machine learning that uses or relates information from different modalities, such as language, sound, images, touch or sensor...
Neocognitron - The neocognitron is a hierarchical neural-network model for visual pattern recognition introduced by Kunihiko Fukushima.
Neural Architecture Search - Neural architecture search, or NAS, is the use of automated methods to explore neural-network designs and select promising architectures for a task.
Neural cryptography - Neural cryptography is research into the use of neural-network methods in cryptographic tasks and their analysis.
Neural field - In machine learning, a neural field is a representation of a quantity over coordinates in which a neural network supplies all or part of the mapping.
Neural gas - Neural gas is an unsupervised learning method that represents a data distribution using movable prototype vectors.
Neural network - A neural network is a computational model made from connected units that transform numerical information.
Neural network Gaussian process - A neural network Gaussian process, or NNGP, is a Gaussian-process model associated with the infinite-width limit of certain randomly initialized neural...
Neural network pruning - Neural network pruning removes or disables selected connections, parameters, or structural components of a neural network.
Neural operators - Neural operators are neural-network models designed to learn mappings between functions.
Neural style transfer - Neural style transfer is a family of image-transformation methods that uses neural representations to combine aspects of one image's content with aspects of...
Neural Turing Machine Network - A Neural Turing Machine, or NTM, is a neural architecture that combines a trainable controller with an external memory matrix.
Neuro-fuzzy systems - Neuro-fuzzy systems combine neural or adaptive learning methods with fuzzy inference.
node2vec - node2vec is a graph representation method that learns a numerical vector for each node by predicting neighborhood relationships sampled through biased...
Oja's rule - Oja's rule is an unsupervised learning rule for a linear neuron that combines Hebbian reinforcement with an activity-dependent stabilizing term.
Optical neural network - An optical neural network performs some or all of its neural computation using light and optical components.
Oscillatory neural network - An oscillatory neural network is a computational network whose units have oscillating states and whose interactions are used to process information.
Output layer - The output layer is the part of a neural network that produces its externally used prediction or representation.
Outstar - An outstar is a neural learning arrangement in which one source unit sends adaptable connections to several target units.
Perceptron - A perceptron is a simple linear classification model that combines numerical inputs with weights and a bias, then applies a threshold to choose a class.
Perceptron learning rule - The perceptron learning rule is an error-driven procedure for adjusting a linear classifier's weights and bias.
Physical neural network - A physical neural network uses controllable physical processes to carry out neural computation.
Physics-informed neural networks - Physics-informed neural networks, or PINNs, are neural models trained using physical equations or related scientific constraints as well as available...
Planning and control layer - The planning and control layer is the part of a robot architecture that converts goals and an estimated world state into planned motion and executable...
Pooling Layer - A pooling layer summarizes groups of values in a neural-network representation.
Preprocessing - Preprocessing is the preparation and transformation of data before it is used by a model or another analysis step.
Probabilistic Topic Model - A probabilistic topic model is a statistical model that represents a collection of documents through latent themes and probability distributions.
Promoter based genetic algorithm - The promoter based genetic algorithm, or PBGA, is a neuroevolution method that uses promoter-like switches to control which encoded neural units are...
Proportional Hazards Model - A proportional hazards model describes how characteristics multiply the instantaneous rate of an event among cases that have not yet experienced it.
Pulse-coupled networks - Pulse-coupled networks are networks of interacting units that influence one another through pulses.
Quantum neural network - A quantum neural network is a learning model that uses quantum operations in a network-like or trainable computational structure.
Radial Basis Network - A radial basis network is a neural network that represents inputs through functions of their distance from selected centers.
Random Effects Model - A random effects model represents selected differences between groups, people, devices or other units as draws from a shared distribution.
Random neural network - The random neural network is Erol Gelenbe's stochastic neural model in which units hold non-negative integer potentials and exchange excitatory or...
Reactive layer - A reactive layer is the part of a robot or agent architecture that produces timely responses to current observations through relatively direct...
Receptron - A receptron is a proposed nonlinear threshold computing unit that generalizes the perceptron by allowing effective input weights to depend on the input...
Recurrent Layer - A recurrent layer is a neural-network component that processes a sequence by updating an internal state as it encounters successive inputs.
Recurrent neural network - A recurrent neural network, or RNN, is a neural-network architecture that processes sequential information using a state updated through repeated computations.
Recursive Neural Network - A recursive neural network is a neural model that repeatedly combines representations of smaller components to represent a larger structured object.
Region Based Convolutional Neural Networks - Region-based convolutional neural networks are object-detection approaches that analyze candidate image regions using convolutional features to identify...
Relation network - A relation network is a neural-network module designed to learn from relationships between pairs of represented objects.
Reservoir computing - Reservoir computing is an approach to processing time-dependent information in which an input drives a dynamical system, called a reservoir, and a readout...