Locally weighted learning - Locally weighted learning makes a prediction by emphasizing training examples relevant to the query being answered.
Logistic loss - Logistic loss is a smooth loss function for binary prediction that penalizes confidence assigned to the wrong outcome.
Long-distance dependencies - Long-distance dependencies are relationships in which interpreting or predicting one part of a sequence requires information from a substantially separated...
Longitudinal data - Longitudinal data contain repeated observations of the same entities over time. The entities might be people, machines, stores, or accounts.
LU decomposition - LU decomposition factors a matrix into a lower-triangular matrix L and an upper-triangular matrix U, often after row permutations.
Mahalanobis distance - Mahalanobis distance measures separation between numerical vectors after accounting for a chosen covariance structure.
Majorize-minimize - Majorize-minimize, or MM, is an optimization principle that repeatedly minimizes an upper-bounding surrogate for the objective.
Manhattan distance - Manhattan distance is the sum of the absolute coordinate differences between two numerical vectors.
Marginalization - Marginalization is the process of obtaining the probability distribution of selected variables from a joint distribution by summing or integrating over the...
Markov Chain - A Markov chain is a sequence of random states in which the conditional distribution of the next state depends on the past only through the current state.
Martingale (probability theory) - A martingale is a stochastic process whose next value has conditional expectation equal to its current value, given the information available so far.
Matrix - A matrix is a rectangular array of entries arranged in rows and columns. A matrix with m rows and n columns has shape m-by-n.
Matrix calculus - Matrix calculus is the study of derivatives of functions whose inputs or outputs are vectors and matrices.
Matrix chain multiplication - Matrix chain multiplication is the problem of choosing how to place parentheses in an ordered product of compatible matrices so that evaluating the product...
Matrix Chernoff bound - A matrix Chernoff bound is a probability inequality controlling extreme eigenvalues of a sum of independent, bounded, positive-semidefinite random matrices.
Matrix difference equation - A matrix difference equation describes how a vector or matrix changes from one discrete step to the next using matrix operations.
Matrix factorization - Matrix factorization expresses a matrix as a product of simpler matrices, either exactly or approximately.
Matrix norm - A matrix norm is a rule that assigns a nonnegative measure of size to a matrix, satisfying the usual norm properties: only the zero matrix has size zero...
Matrix sign function - The matrix sign function maps a square matrix's spectral components in the open right half of the complex plane to +1 and those in the open left half to −1.
Max norm - The max norm of a finite-dimensional vector is the largest absolute value of any of its coordinates.
Maximal entropy random walk - A maximal entropy random walk is a graph-based Markov process whose transition probabilities maximize the long-run entropy rate of allowed paths.
Maximum-entropy Markov model - A maximum-entropy Markov model, or MEMM, is a conditional sequence-labeling model that uses a log-linear classifier to predict each label from the preceding...
Maximum entropy spectral estimation - Maximum entropy spectral estimation estimates how a stationary time series distributes power across frequencies by choosing a spectrum with maximum entropy...
MCS algorithm - The MCS algorithm, in mathematical optimization, is Multilevel Coordinate Search: a method for minimizing a function subject to coordinate bounds using...
Measure space - A measure space is a mathematical structure consisting of a set, a collection of its measurable subsets, and a rule assigning a nonnegative size to each...
Mehrotra predictor - corrector method - Mehrotra's predictor - corrector method is a primal - dual interior-point optimization method that combines an affine-scaling prediction with a correction...
Mercer's theorem - Mercer's theorem is a result explaining how suitable continuous, symmetric, positive-semidefinite kernels can be represented by nonnegative eigenvalues and...
Message passing - Message passing is a way for computational participants to communicate by sending and receiving messages.
Method of moving asymptotes - The method of moving asymptotes, or MMA, is a gradient-based method for constrained nonlinear optimization.
Method of multipliers - The method of multipliers is an iterative method for constrained optimization that alternates between minimizing an augmented Lagrangian and updating the...
Method of types - The method of types is an information-theoretic technique that groups finite sequences according to their empirical symbol frequencies.
Min-max optimization - Min-max optimization chooses a decision that minimizes the largest value of an objective over a specified set of alternatives, scenarios, or adversarial...
Mini-batch stochastic gradient descent - Mini-batch stochastic gradient descent is an optimization method that updates model parameters using gradients calculated from a relatively small group of...
Minimax approximation algorithm - A minimax approximation algorithm finds an approximation that minimizes the largest error over a specified domain, within a chosen family of approximating...
Minimum edit distance - Minimum edit distance is the smallest total cost of editing operations needed to transform one sequence into another.
Minkowski distance - Minkowski distance is a family of distances between numerical vectors, defined for p≥1 by d_p(x,y)=(sum_i |x_i−y_i|^p)^(1/p).
Mirror descent - Mirror descent is an optimization method that generalizes gradient descent by using a chosen geometry to control how far each update moves.
Missing data - Missing data are values that should be available for an observation but are absent, unknown, corrupted, or otherwise unusable.
Mixed integer linear programming - Mixed integer linear programming, usually abbreviated MILP, is an optimization framework with a linear objective, linear constraints, and integer...
Model order reduction - Model order reduction constructs a lower-dimensional model that approximates important behavior of a larger dynamical system.
Moment generating function - A moment generating function, or MGF, represents a random variable X through the function M_X(t)=E[exp(tX)], wherever this expectation is finite.
MRF optimization via dual decomposition - MRF optimization via dual decomposition is a framework for minimizing a discrete Markov random field's energy by splitting it into easier subproblems and...
Multi-attribute global inference of quality - Multi-attribute global inference of quality, usually abbreviated MAGIQ, is a decision-analysis method that combines rankings on several attributes into an...
Multi-objective optimization - Multi-objective optimization studies decisions with two or more objectives that may conflict, such as reducing prediction error and reducing computing cost.
Multicriteria classification - Multicriteria classification is the assignment of alternatives to predefined categories using several evaluation criteria.
Multigrid method - A multigrid method solves a numerical problem by combining calculations at several levels of resolution.
Multilabel indicator matrix - A multilabel indicator matrix is a two-dimensional representation of labels in which each row corresponds to a sample and each column corresponds to a...
Multilevel Monte Carlo method - The multilevel Monte Carlo method estimates an expected quantity by combining many inexpensive coarse simulations with fewer carefully paired corrections...