The concept of Gaussian processes is named after Carl Friedrich Gauss because it is based on the notion of the Gaussian distribution (normal distribution). Gaussian processes can be seen as an infinite-dimensional generalization of multivariate normal distributions.
Gaussian processes are useful in statistical modelling, benefiting from properties inherited from the normal distribution. For example, if a random process is modelled as a Gaussian process, the distributions of various derived quantities can be obtained explicitly. Such quantities include the average value of the process over a range of times and the error in estimating the average using sample values at a small set of times. While exact models often scale poorly as the amount of data increases, multiple approximation methods have been developed which often retain good accuracy while drastically reducing computation time.
Using characteristic functions of random variables, the Gaussian property can be formulated as follows: is Gaussian if and only if, for every finite set of indices , there are real-valued , with such that the following equality holds for all
The numbers and can be shown to be the covariances and means of the variables in the process.
The variance of a Gaussian process is finite at any time , formally: p. 515
For general stochastic processes strict-sense stationarity implies wide-sense stationarity but not every wide-sense stationary stochastic process is strict-sense stationary. However, for a Gaussian stochastic process the two concepts are equivalent.: p. 518
A Gaussian stochastic process is strict-sense stationary if, and only if, it is wide-sense stationary.
There is an explicit representation for stationary Gaussian processes. A simple example of this representation is
A key fact of Gaussian processes is that they can be completely defined by their second-order statistics. Thus, if a Gaussian process is assumed to have mean zero, defining the covariance function completely defines the process' behaviour. Importantly the non-negative definiteness of this function enables its spectral decomposition using the Karhunen–Loève expansion. Basic aspects that can be defined through the covariance function are the process' stationarity, isotropy, smoothness and periodicity.
Stationarity refers to the process' behaviour regarding the separation of any two points and . If the process is stationary, the covariance function depends only on . For example, the Ornstein–Uhlenbeck process is stationary.
If the process depends only on , the Euclidean distance (not the direction) between and , then the process is considered isotropic. A process that is concurrently stationary and isotropic is considered to be homogeneous; in practice these properties reflect the differences (or rather the lack of them) in the behaviour of the process given the location of the observer.
Ultimately Gaussian processes translate as taking priors on functions and the smoothness of these priors can be induced by the covariance function. If we expect that for "near-by" input points and their corresponding output points and to be "near-by" also, then the assumption of continuity is present. If we wish to allow for significant displacement then we might choose a rougher covariance function. Extreme examples of the behaviour is the Ornstein–Uhlenbeck covariance function and the squared exponential where the former is never differentiable and the latter infinitely differentiable.
Periodicity refers to inducing periodic patterns within the behaviour of the process. Formally, this is achieved by mapping the input to a two dimensional vector .
Usual covariance functions
The effect of choosing different kernels on the prior function distribution of the Gaussian process. Left is a squared exponential kernel. Middle is Brownian. Right is quadratic.
There are a number of common covariance functions:
white Gaussian noise:
Here . The parameter is the characteristic length-scale of the process (practically, "how close" two points and have to be to influence each other significantly), is the Kronecker delta and the standard deviation of the noise fluctuations. Moreover, is the modified Bessel function of order and is the gamma function evaluated at . Importantly, a complicated covariance function can be defined as a linear combination of other simpler covariance functions in order to incorporate different insights about the data-set at hand.
The inferential results are dependent on the values of the hyperparameters (e.g. and ) defining the model's behaviour. A popular choice for is to provide maximum a posteriori (MAP) estimates of it with some chosen prior. If the prior is very near uniform, this is the same as maximizing the marginal likelihood of the process; the marginalization being done over the observed process values . This approach is also known as maximum likelihood II, evidence maximization, or empirical Bayes.
For a stationary Gaussian process some conditions on its spectrum are sufficient for sample continuity, but fail to be necessary. A necessary and sufficient condition, sometimes called Dudley–Fernique theorem, involves the function defined by
(the right-hand side does not depend on due to stationarity). Continuity of in probability is equivalent to continuity of at When convergence of to (as ) is too slow, sample continuity of may fail. Convergence of the following integrals matters:
these two integrals being equal according to integration by substitution The first integrand need not be bounded as thus the integral may converge () or diverge (). Taking for example for large that is, for small one obtains when and when
In these two cases the function is increasing on but generally it is not. Moreover, the condition
(*) there exists such that is monotone on
does not follow from continuity of and the evident relations (for all ) and
Theorem 1 — Let be continuous and satisfy (*). Then the condition is necessary and sufficient for sample continuity of
There exist sample continuous processes such that they violate condition (*). An example found by Marcus and Shepp : 387 is a random lacunary Fourier series
where are independent random variables with standard normal distribution; frequencies are a fast growing sequence; and coefficients satisfy The latter relation implies whence almost surely, which ensures uniform convergence of the Fourier series almost surely, and sample continuity of
Autocorrelation of a random lacunary Fourier series
Its autocovariation function
is nowhere monotone (see the picture), as well as the corresponding function
Brownian motion as the integral of Gaussian processes
As such, almost all sample paths of a mean-zero Gaussian process with positive definite kernel will lie outside of the Hilbert space .
Linearly constrained Gaussian processes
For many applications of interest some pre-existing knowledge about the system at hand is already given. Consider e.g. the case where the output of the Gaussian process corresponds to a magnetic field; here, the real magnetic field is bound by Maxwell's equations and a way to incorporate this constraint into the Gaussian process formalism would be desirable as this would likely improve the accuracy of the algorithm.
A method on how to incorporate linear constraints into Gaussian processes already exists:
Consider the (vector valued) output function which is known to obey the linear constraint (i.e. is a linear operator)
Then the constraint can be fulfilled by choosing , where is modelled as a Gaussian process, and finding such that
Given and using the fact that Gaussian processes are closed under linear transformations, the Gaussian process for obeying constraint becomes
Hence, linear constraints can be encoded into the mean and covariance function of a Gaussian process.
An example of Gaussian Process Regression (prediction) compared with other regression models.
A Gaussian process can be used as a prior probability distribution over functions in Bayesian inference. Given any set of N points in the desired domain of your functions, take a multivariate Gaussian whose covariance matrix parameter is the Gram matrix of your N points with some desired kernel, and sample from that Gaussian. For solution of the multi-output prediction problem, Gaussian process regression for vector-valued function was developed. In this method, a 'big' covariance is constructed, which describes the correlations between all the input and output variables taken in N points in the desired domain. This approach was elaborated in detail for the matrix-valued Gaussian processes and generalised to processes with 'heavier tails' like Student-t processes.
Inference of continuous values with a Gaussian process prior is known as Gaussian process regression, or kriging; extending Gaussian process regression to multiple target variables is known as cokriging. Gaussian processes are thus useful as a powerful non-linear multivariate interpolation tool.
Gaussian processes are also commonly used to tackle numerical analysis problems such as numerical integration, solving differential equations, or optimisation in the field of probabilistic numerics.
Gaussian processes can also be used in the context of mixture of experts models, for example. The underlying rationale of such a learning framework consists in the assumption that a given mapping cannot be well captured by a single Gaussian process model. Instead, the observation space is divided into subsets, each of which is characterized by a different mapping function; each of these is learned via a different Gaussian process component in the postulated mixture.
Gaussian Process Regression (prediction) with a squared exponential kernel. Left plot are draws from the prior function distribution. Middle are draws from the posterior. Right is mean prediction with one standard deviation shaded.
When concerned with a general Gaussian process regression problem (Kriging), it is assumed that for a Gaussian process observed at coordinates , the vector of values is just one sample from a multivariate Gaussian distribution of dimension equal to number of observed coordinates . Therefore, under the assumption of a zero-mean distribution, , where is the covariance matrix between all possible pairs for a given set of hyperparameters θ.
As such the log marginal likelihood is:
and maximizing this marginal likelihood towards θ provides the complete specification of the Gaussian process f. One can briefly note at this point that the first term corresponds to a penalty term for a model's failure to fit observed values and the second term to a penalty term that increases proportionally to a model's complexity. Having specified θ, making predictions about unobserved values at coordinates x* is then only a matter of drawing samples from the predictive distribution where the posterior mean estimate A is defined as
and the posterior variance estimate B is defined as:
where is the covariance between the new coordinate of estimation x* and all other observed coordinates x for a given hyperparameter vector θ, and are defined as before and is the variance at point x* as dictated by θ. It is important to note that practically the posterior mean estimate of (the "point estimate") is just a linear combination of the observations ; in a similar manner the variance of is actually independent of the observations . A known bottleneck in Gaussian process prediction is that the computational complexity of inference and likelihood evaluation is cubic in the number of points |x|, and as such can become unfeasible for larger data sets. Works on sparse Gaussian processes, that usually are based on the idea of building a representative set for the given process f, try to circumvent this issue. The kriging method can be used in the latent level of a nonlinear mixed-effects model for a spatial functional prediction: this technique is called the latent kriging.
Often, the covariance has the form , where is a scaling parameter. Examples are the Matérn class covariance functions. If this scaling parameter is either known or unknown (i.e. must be marginalized), then the posterior probability, , i.e. the probability for the hyperparameters given a set of data pairs of observations of and , admits an analytical expression.
Bayesian neural networks are a particular type of Bayesian network that results from treating deep learning and artificial neural network models probabilistically, and assigning a prior distribution to their parameters. Computation in artificial neural networks is usually organized into sequential layers of artificial neurons. The number of neurons in a layer is called the layer width. As layer width grows large, many Bayesian neural networks reduce to a Gaussian process with a closed form compositional kernel. This Gaussian process is called the Neural Network Gaussian Process (NNGP). It allows predictions from Bayesian neural networks to be more efficiently evaluated, and provides an analytic tool to understand deep learning models.
In practical applications, Gaussian process models are often evaluated on a grid leading to multivariate normal distributions. Using these models for prediction or parameter estimation using maximum likelihood requires evaluating a multivariate Gaussian density, which involves calculating the determinant and the inverse of the covariance matrix. Both of these operations have cubic computational complexity which means that even for grids of modest sizes, both operations can have a prohibitive computational cost. This drawback led to the development of multiple approximation methods.
^Ledoux, Michel (1996), "Isoperimetry and Gaussian analysis", in Dobrushin, Roland; Groeneboom, Piet; Ledoux, Michel (eds.), Lectures on Probability Theory and Statistics: Ecole d'Eté de Probabilités de Saint-Flour XXIV–1994, Lecture Notes in Mathematics, vol. 1648, Berlin: Springer, pp. 165–294, doi:10.1007/BFb0095676, MR1600888
^Adler, Robert J. (1990). An Introduction to Continuity, Extrema, and Related Topics for General Gaussian Processes. Lecture Notes-Monograph Series. Institute of Mathematical Statistics Lecture Notes—Monograph Series. Vol. 12. Hayward, California: Institute of Mathematical Statistics. ISBN0-940600-17-X. JSTOR4355563. MR1088478.
^Berman, Simeon M. (1992). "Review of: Adler 1990 'An introduction to continuity...'". Mathematical Reviews. MR1088478.
^Driscoll, Michael F. (1973). "The reproducing kernel Hilbert space structure of the sample paths of a Gaussian process". Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete. 26 (4): 309–316. doi:10.1007/BF00534894. ISSN0044-3719. S2CID123348980.
^Jidling, Carl; Wahlström, Niklas; Wills, Adrian; Schön, Thomas B. (2017-09-19). "Linearly constrained Gaussian processes". arXiv:1703.00787 [stat.ML].
^Chatzis, Sotirios P. (2013). "A latent variable Gaussian process model with Pitman–Yor process priors for multiclass classification". Neurocomputing. 120: 482–489. doi:10.1016/j.neucom.2013.04.029.
^Smola, A.J.; Schoellkopf, B. (2000). "Sparse greedy matrix approximation for machine learning". Proceedings of the Seventeenth International Conference on Machine Learning: 911–918. CiteSeerX10.1.1.43.3153.
^Novak, Roman; Xiao, Lechao; Hron, Jiri; Lee, Jaehoon; Alemi, Alexander A.; Sohl-Dickstein, Jascha; Schoenholz, Samuel S. (2020). "Neural Tangents: Fast and Easy Infinite Neural Networks in Python". International Conference on Learning Representations. arXiv:1912.02803.
^Neal, Radford M. (2012). Bayesian Learning for Neural Networks. Springer Science and Business Media.