In probability theory and mathematical physics, a random matrix is a matrixvalued random variable—that is, a matrix in which some or all elements are random variables. Many important properties of physical systems can be represented mathematically as matrix problems. For example, the thermal conductivity of a lattice can be computed from the dynamical matrix of the particleparticle interactions within the lattice.
Random matrix theory can be applied to the electrical and communications engineering research efforts to study, model and develop Massive MultipleInput MultipleOutput (MIMO) radio systems.
In nuclear physics, random matrices were introduced by Eugene Wigner to model the nuclei of heavy atoms.^{[1]} Wigner postulated that the spacings between the lines in the spectrum of a heavy atom nucleus should resemble the spacings between the eigenvalues of a random matrix, and should depend only on the symmetry class of the underlying evolution.^{[2]} In solidstate physics, random matrices model the behaviour of large disordered Hamiltonians in the meanfield approximation.
In quantum chaos, the Bohigas–Giannoni–Schmit (BGS) conjecture asserts that the spectral statistics of quantum systems whose classical counterparts exhibit chaotic behaviour are described by random matrix theory.^{[3]}
In quantum optics, transformations described by random unitary matrices are crucial for demonstrating the advantage of quantum over classical computation (see, e.g., the boson sampling model).^{[4]} Moreover, such random unitary transformations can be directly implemented in an optical circuit, by mapping their parameters to optical circuit components (that is beam splitters and phase shifters).^{[5]}
Random matrix theory has also found applications to the chiral Dirac operator in quantum chromodynamics,^{[6]} quantum gravity in two dimensions,^{[7]} mesoscopic physics,^{[8]} spintransfer torque,^{[9]} the fractional quantum Hall effect,^{[10]} Anderson localization,^{[11]} quantum dots,^{[12]} and superconductors^{[13]}
In multivariate statistics, random matrices were introduced by John Wishart, who sought to estimate covariance matrices of large samples.^{[14]} Chernoff, Bernstein, and Hoeffdingtype inequalities can typically be strengthened when applied to the maximal eigenvalue (i.e. the eigenvalue of largest magnitude) of a finite sum of random Hermitian matrices.^{[15]} Random matrix theory is used to study the spectral properties of random matrices—such as sample covariance matrices—which is of particular interest in highdimensional statistics. Random matrix theory also saw applications in neuronal networks^{[16]} and deep learning, with recent work utilizing random matrices to show that hyperparameter tunings can be cheaply transferred between large neural networks without the need for retraining.^{[17]}
In numerical analysis, random matrices have been used since the work of John von Neumann and Herman Goldstine^{[18]} to describe computation errors in operations such as matrix multiplication. Although random entries are traditional "generic" inputs to an algorithm, the concentration of measure associated with random matrix distributions implies that random matrices will not test large portions of an algorithm's input space.^{[19]}
In number theory, the distribution of zeros of the Riemann zeta function (and other Lfunctions) is modeled by the distribution of eigenvalues of certain random matrices.^{[20]} The connection was first discovered by Hugh Montgomery and Freeman Dyson. It is connected to the Hilbert–Pólya conjecture.
The relation of free probability with random matrices^{[21]} is a key reason for the wide use of free probability in other subjects. Voiculescu introduced the concept of freeness around 1983 in an operator algebraic context; at the beginning there was no relation at all with random matrices. This connection was only revealed later in 1991 by Voiculescu;^{[22]} he was motivated by the fact that the limit distribution which he found in his free central limit theorem had appeared before in Wigner's semicircle law in the random matrix context.
In the field of computational neuroscience, random matrices are increasingly used to model the network of synaptic connections between neurons in the brain. Dynamical models of neuronal networks with random connectivity matrix were shown to exhibit a phase transition to chaos^{[23]} when the variance of the synaptic weights crosses a critical value, at the limit of infinite system size. Results on random matrices have also shown that the dynamics of randommatrix models are insensitive to mean connection strength. Instead, the stability of fluctuations depends on connection strength variation^{[24]}^{[25]} and time to synchrony depends on network topology.^{[26]}^{[27]}
In the analysis of massive data such as fMRI, random matrix theory has been applied in order to perform dimension reduction. When applying an algorithm such as PCA, it is important to be able to select the number of significant components. The criteria for selecting components can be multiple (based on explained variance, Kaiser's method, eigenvalue, etc.). Random matrix theory in this content has its representative the MarchenkoPastur distribution, which guarantees the theoretical high and low limits of the eigenvalues associated with a random variable covariance matrix. This matrix calculated in this way becomes the null hypothesis that allows one to find the eigenvalues (and their eigenvectors) that deviate from the theoretical random range. The components thus excluded become the reduced dimensional space (see examples in fMRI ^{[28]}^{[29]}).
In optimal control theory, the evolution of n state variables through time depends at any time on their own values and on the values of k control variables. With linear evolution, matrices of coefficients appear in the state equation (equation of evolution). In some problems the values of the parameters in these matrices are not known with certainty, in which case there are random matrices in the state equation and the problem is known as one of stochastic control.^{[30]}^{: ch. 13 }^{[31]}^{[32]} A key result in the case of linearquadratic control with stochastic matrices is that the certainty equivalence principle does not apply: while in the absence of multiplier uncertainty (that is, with only additive uncertainty) the optimal policy with a quadratic loss function coincides with what would be decided if the uncertainty were ignored, the optimal policy may differ if the state equation contains random coefficients.
In computational mechanics, epistemic uncertainties underlying the lack of knowledge about the physics of the modeled system give rise to mathematical operators associated with the computational model, which are deficient in a certain sense. Such operators lack certain properties linked to unmodeled physics. When such operators are discretized to perform computational simulations, their accuracy is limited by the missing physics. To compensate for this deficiency of the mathematical operator, it is not enough to make the model parameters random, it is necessary to consider a mathematical operator that is random and can thus generate families of computational models in the hope that one of these captures the missing physics. Random matrices have been used in this sense,^{[33]}^{[34]} with applications in vibroacoustics, wave propagations, materials science, fluid mechanics, heat transfer, etc.
The mostcommonly studied random matrix distributions are the Gaussian ensembles: GOE, GUE and GSE. They are often denoted by their Dyson index, β = 1 for GOE, β = 2 for GUE, and β = 4 for GSE. This index counts the number of real components per matrix element.
The Gaussian unitary ensemble is described by the Gaussian measure with density
The Gaussian orthogonal ensemble is described by the Gaussian measure with density
The Gaussian symplectic ensemble is described by the Gaussian measure with density
The ensembles as defined here have Gaussian distributed matrix elements with mean ⟨H_{ij}⟩ = 0, and twopoint correlations given by
The moment generating function for the GOE is
The joint probability density for the eigenvalues λ_{1}, λ_{2}, ..., λ_{n} of GUE/GOE/GSE is given by

(1) 
where Z_{β,n} is a normalization constant which can be explicitly computed, see Selberg integral. In the case of GUE (β = 2), the formula (1) describes a determinantal point process. Eigenvalues repel as the joint probability density has a zero (of th order) for coinciding eigenvalues .
The distribution of the largest eigenvalue for GOE, and GUE, are explicitly solvable.^{[35]} They converge to the Tracy–Widom distribution after shifting and scaling appropriately.
The spectrum, divided by , converges in distribution to the semicircular distribution on the interval : . Here is the variance of offdiagonal entries.
From the ordered sequence of eigenvalues , one defines the normalized spacings , where is the mean spacing. The probability distribution of spacings is approximately given by,
The numerical constants are such that is normalized:
Wigner matrices are random Hermitian matrices such that the entries
Invariant matrix ensembles are random Hermitian matrices with density on the space of real symmetric/Hermitian/quaternionic Hermitian matrices, which is of the form where the function V is called the potential.
The Gaussian ensembles are the only common special cases of these two classes of random matrices. This is a consequence of a theorem by Porter and Rosenzweig.^{[36]}^{[37]}
The spectral theory of random matrices studies the distribution of the eigenvalues as the size of the matrix goes to infinity.
In the global regime, one is interested in the distribution of linear statistics of the form .
The empirical spectral measure μ_{H} of H is defined by
Usually, the limit of is a deterministic measure; this is a particular case of selfaveraging. The cumulative distribution function of the limiting measure is called the integrated density of states and is denoted N(λ). If the integrated density of states is differentiable, its derivative is called the density of states and is denoted ρ(λ).
The limit of the empirical spectral measure for Wigner matrices was described by Eugene Wigner; see Wigner semicircle distribution and Wigner surmise. As far as sample covariance matrices are concerned, a theory was developed by Marčenko and Pastur.^{[38]}^{[39]}
The limit of the empirical spectral measure of invariant matrix ensembles is described by a certain integral equation which arises from potential theory.^{[40]}
For the linear statistics N_{f,H} = n^{−1} Σ f(λ_{j}), one is also interested in the fluctuations about ∫ f(λ) dN(λ). For many classes of random matrices, a central limit theorem of the form
Consider the measure
where is the potential of the ensemble and let be the empirical spectral measure.
We can rewrite with as
the probability measure is now of the form
where is the above functional inside the squared brackets.
Let now
be the space of onedimensional probability measures and consider the minimizer
For there exists a unique equilibrium measure through the EulerLagrange variational conditions for some real constant
where is the support of the measure and
The equilibrium measure has the following Radon–Nikodym density
In the local regime, one is interested in the spacings between eigenvalues, and, more generally, in the joint distribution of eigenvalues in an interval of length of order 1/n. One distinguishes between bulk statistics, pertaining to intervals inside the support of the limiting spectral measure, and edge statistics, pertaining to intervals near the boundary of the support.
Formally, fix in the interior of the support of . Then consider the point process
The point process captures the statistical properties of eigenvalues in the vicinity of . For the Gaussian ensembles, the limit of is known;^{[2]} thus, for GUE it is a determinantal point process with the kernel
The universality principle postulates that the limit of as should depend only on the symmetry class of the random matrix (and neither on the specific model of random matrices nor on ). Rigorous proofs of universality are known for invariant matrix ensembles^{[44]}^{[45]} and Wigner matrices.^{[46]}^{[47]}
Main article: Tracy–Widom distribution 
^{[48]} The typical statement of the Wigner semicircular law is equivalent to the following statement: For each fixed interval centered at a point , as , the number of dimensions of the gaussian ensemble increases, the proportion of the eigenvalues falling within the interval converges to , where is the density of the semicircular distribution.
If can be allowed to decrease as increases, then we obtain strictly stronger theorems, named "local laws".
The joint probability density of the eigenvalues of random Hermitian matrices , with partition functions of the form
The following result expresses these correlation functions as determinants of the matrices formed from evaluating the appropriate integral kernel at the pairs of points appearing within the correlator.
Theorem [DysonMehta] For any , the point correlation function can be written as a determinant
Main article: Wishart distribution 
Wishart matrices are n × n random matrices of the form H = X X^{*}, where X is an n × m random matrix (m ≥ n) with independent entries, and X^{*} is its conjugate transpose. In the important special case considered by Wishart, the entries of X are identically distributed Gaussian random variables (either real or complex).
The limit of the empirical spectral measure of Wishart matrices was found^{[38]} by Vladimir Marchenko and Leonid Pastur.
Main article: Circular ensembles 
Main article: Circular law 