Posterior Gaussian process samples by updating prior samples

Matheron’s other weird trick

December 3, 2020 — September 20, 2022

Figure 1

\[\renewcommand{\var}{\operatorname{Var}} \renewcommand{\cov}{\operatorname{Cov}} \renewcommand{\corr}{\operatorname{Corr}} \renewcommand{\dd}{\mathrm{d}} \renewcommand{\vv}[1]{\boldsymbol{#1}} \renewcommand{\rv}[1]{\mathsf{#1}} \renewcommand{\vrv}[1]{\vv{\rv{#1}}} \renewcommand{\disteq}{\stackrel{d}{=}} \renewcommand{\dif}{\backslash} \renewcommand{\gvn}{\mid} \renewcommand{\Ex}{\mathbb{E}} \renewcommand{\Pr}{\mathbb{P}}\]

Can we find a transformation that will turn a Gaussian process prior sample into a Gaussian process posterior sample. A special trick where we do GP regression by GP simulation.

The main tool is an old insight made useful for modern problems in J. T. Wilson et al. (2020) (brusque) and J. T. Wilson et al. (2021) (deep). Actioned in Ritter et al. (2021) to condition probabilistic neural nets somehow.

Danger: notation updates in the pipeline.

1 Matheron updates for Gaussian RVs

We start by examining a slightly different way of defining a Gaussian RV (J. T. Wilson et al. 2021) starting from the recipe for sampling:

A random vector \(\boldsymbol{x}=\left(x_{1}, \ldots, x_{n}\right) \in \mathbb{R}^{n}\) is said to be Gaussian if there exists a matrix \(\mathbf{L}\) and vector \(\boldsymbol{\mu}\) such that \[ \boldsymbol{x} \stackrel{\mathrm{d}}{=} \boldsymbol{\mu}+\mathbf{L} \boldsymbol{\zeta} \quad \boldsymbol{\zeta} \sim \mathcal{N}(\mathbf{0}, \mathbf{I}) \] where \(\mathcal{N}(\mathbf{0}, \mathbf{I})\) is known as the standard version of a (multivariate) normal distribution, which is defined through its density.

This is the location-scale form of a Gaussian RV, as opposed to the canonical form which we use in Gaussian Belief Propagation. In location-scale form, a non-degenerate Gaussian RV’s distribution is given (uniquely) by its mean \(\boldsymbol{\mu}=\mathbb{E}(\boldsymbol{x})\) and its covariance \(\boldsymbol{\Sigma}=\mathbb{E}\left[(\boldsymbol{x}-\boldsymbol{\mu})(\boldsymbol{x}-\boldsymbol{\mu})^{\top}\right] .\) In this notation the density, if defined, is \[ p(\boldsymbol{x})=\mathcal{N}(\boldsymbol{x} ; \boldsymbol{\mu}, \boldsymbol{\Sigma})=\frac{1}{\sqrt{|2 \pi \boldsymbol{\Sigma}|}} \exp \left(-\frac{1}{2}(\boldsymbol{x}-\boldsymbol{\mu})^{\top} \boldsymbol{\Sigma}^{-1}(\boldsymbol{x}-\boldsymbol{\mu})\right). \]

Since \(\zeta\) has identity covariance, any matrix square root of \(\boldsymbol{\Sigma}\), such as the Cholesky factor \(\mathbf{L}\) with \(\boldsymbol{\Sigma}=\mathbf{L L}^{\top}\), may be used to draw \(\boldsymbol{x}=\boldsymbol{\mu}+\mathbf{L} \boldsymbol{\zeta}.\)

tl;dr we can think about drawing any Gaussian RV as transforming a standard Gaussian. So much is basic entry-level stuff. What might a rule which updates a Gaussian prior into a data-conditioned posterior look like? Like this!

We define \(\cov(a,b)=\Sigma_{a,b}\) as the covariance between two random variables (J. T. Wilson et al. 2021):

Matheron’s Update Rule: Let \(\boldsymbol{a}\) and \(\boldsymbol{b}\) be jointly Gaussian, centered random variables. Then the random variable \(\boldsymbol{a}\) conditional on \(\boldsymbol{b}=\boldsymbol{\beta}\) may be expressed as \[ (\boldsymbol{a} \mid \boldsymbol{b}=\boldsymbol{\beta}) \stackrel{\mathrm{d}}{=} \boldsymbol{a}+\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{b}}{\boldsymbol{\Sigma}}_{\boldsymbol{b}, \boldsymbol{b}}^{-1}(\boldsymbol{\beta}-\boldsymbol{b}) \] Proof: Comparing the mean and covariance on both sides immediately affirms the result \[ \begin{aligned} \mathbb{E}\left(\boldsymbol{a}+\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{b}} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{b}}^{-1}(\boldsymbol{\beta}-\boldsymbol{b})\right) & =\boldsymbol{\mu}_{\boldsymbol{a}}+\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{b}} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{b}}^{-1}\left(\boldsymbol{\beta}-\boldsymbol{\mu}_{\boldsymbol{b}}\right) \\ & =\mathbb{E}(\boldsymbol{a} \mid \boldsymbol{b}=\boldsymbol{\beta}) \end{aligned} \] \[ \begin{aligned} \operatorname{Cov}\left(\boldsymbol{a}+\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{b}} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{b}}^{-1}(\boldsymbol{\beta}-\boldsymbol{b})\right) &=\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{a}}+\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{b}} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{b}}^{-1} \operatorname{Cov}(\boldsymbol{b}) \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{b}}^{-1} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{a}} \\ & =\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{a}}+\boldsymbol{\Sigma}_{\boldsymbol{a}, \boldsymbol{b}} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{b}}^{-1} \boldsymbol{\Sigma}_{\boldsymbol{b}, \boldsymbol{a}}\\ &=\operatorname{Cov}(\boldsymbol{a} \mid \boldsymbol{b} =\boldsymbol{\beta}) \end{aligned} \]

Figure 2: Visualization of Matheron’s update rule for a bivariate normal distribution with correlation coefficient \(\rho=0.75 .\) Left: Draws from \(p(\boldsymbol{a}, \boldsymbol{b})\) are shown along with the marginal distributions. Right: Matheron’s update rule is used to update samples shown on the left subject to the condition \(\boldsymbol{b}=\boldsymbol{\beta}\). This process is illustrated in full for one a particular draw (J. T. Wilson et al. 2021).

Can we find a transformation that will turn a Gaussian process prior sample (i.e. function) into a Gaussian process posterior sample, and thus use prior samples, which are presumably pretty easy, to create posterior ones, which are often hard. If we evaluate the sampled function at a finite number of points, then we can use the Matheron formula to do precisely that. Sometimes this can even be useful. The resulting algorithm uses tricks from both analytic GP regression and Monte Carlo.

The sample based approximation to this is precisely the Ensemble Kalman Filter.

2 “Exact” updates for Gaussian processes

Figure 3

Exact in the sense that we do not approximate the data. These updates are not exact if our basis function representation is only an approximation to the “true” GP (as with classic GPs) and not exact in the sense that we will be using samples to approximate measures. For now we assume that the observation likelihood is Gaussian.1

For a Gaussian process \(f \sim \mathcal{G P}(\mu, k)\) with marginal \(\boldsymbol{f}_{m}=f(\mathbf{Z})\), the process conditioned on \(\boldsymbol{f}_{m}=\boldsymbol{y}\) admits, in distribution, the representation \[ \underbrace{(f \mid \boldsymbol{y})(\cdot)}_{\text {posterior }} \stackrel{\mathrm{d}}{=} \underbrace{f(\cdot)}_{\text {prior }}+\underbrace{k(\cdot, \mathbf{Z}) \mathbf{K}_{m, m}^{-1}\left(\boldsymbol{y}-\boldsymbol{f}_{m}\right)}_{\text {update }}. \]

If our observations are contaminated by additive i.i.d Gaussian noise, \(\boldsymbol{y}=\boldsymbol{f}_{m} +\boldsymbol{\varepsilon}\) with \(\boldsymbol{\varepsilon}\sim\mathcal{N}(\boldsymbol{0}, \sigma^2\mathbf{I}),\) we find \[ \begin{aligned} &\boldsymbol{f}_{*} \mid \boldsymbol{y} \stackrel{\mathrm{d}}{=} \boldsymbol{f}_{*}+\mathbf{K}_{*, n}\left(\mathbf{K}_{n, n}+\sigma^{2} \mathbf{I}\right)^{-1}(\boldsymbol{y}-\boldsymbol{f}-\boldsymbol{\varepsilon}) \end{aligned} \] When sampling from exact GPs we jointly draw \(\boldsymbol{f}_{*}\) and \(\boldsymbol{f}\) from the prior. Then, we combine \(\boldsymbol{f}\) with noise variates \(\varepsilon \sim \mathcal{N}\left(\mathbf{0}, \sigma^{2} \mathbf{I}\right)\) such that \(\boldsymbol{f}+\varepsilon\) constitutes a draw from the prior distribution of \(\boldsymbol{y}\).

Compare this to the equivalent distributional update from classical GP regression which updates the moments of a distribution, not samples from a path — the formulae are related though:

…the conditional distribution is the Gaussian \(\mathcal{N}\left(\boldsymbol{\mu}_{* \mid y}, \mathbf{K}_{*, * \mid y}\right)\) with moments \[\begin{aligned} \boldsymbol{\mu}_{* \mid \boldsymbol{y}}&=\boldsymbol{\mu}_*+\mathbf{K}_{*, n} \mathbf{K}_{n, n}^{-1}\left(\boldsymbol{y}-\boldsymbol{\mu}_n\right) \\ \mathbf{K}_{*, * \mid \boldsymbol{y}}&=\mathbf{K}_{*, *}-\mathbf{K}_{*, n} \mathbf{K}_{n, n}^{-1} \mathbf{K}_{n, *}\end{aligned} \]

Figure 4: Visual overview for pathwise conditioning of Gaussian processes. Left: The residual \(\boldsymbol{r}=\boldsymbol{y}-\boldsymbol{f}_{n}\) (dashed black) of a draw \(f \sim \mathcal{G} \mathcal{P}(0, k)\), shown in orange, given observations \(\boldsymbol{y}\) (black). Middle: A pathwise update (purple) is constructed by Matheron’s update rule. Right: Prior and update are combined to represent conditional (blue). Empirical moments (light blue) of \(10^{5}\) conditioned paths are compared with those of the model (dashed black). The sample average, which matches the posterior mean, has been omitted for clarity. (J. T. Wilson et al. 2021)

3 Using basis functions

For many purposes we need a basis function representation, a.k.a. the weight-space representation. We assert the GP can be written as a random function comprising basis functions \(\boldsymbol{\phi}=\left(\phi_{1}, \ldots, \phi_{\ell}\right)\) with the Gaussian random weight vector \(w\) so that \[ f^{(w)}(\cdot)=\sum_{i=1}^{\ell} w_{i} \phi_{i}(\cdot) \quad \boldsymbol{w} \sim \mathcal{N}\left(\mathbf{0}, \boldsymbol{\Sigma}_{\boldsymbol{w}}\right). \] \(f^{(w)}\) is a random function satisfying \(\boldsymbol{f}^{(\boldsymbol{w})} \sim \mathcal{N}\left(\mathbf{0}, \boldsymbol{\Phi}_{n} \boldsymbol{\Sigma}_{\boldsymbol{w}} \boldsymbol{\Phi}^{\top}\right)\), where \(\boldsymbol{\Phi}_{n}=\boldsymbol{\phi}(\mathbf{X})\) is a \(|\mathbf{X}| \times \ell\) matrix of features. If we are lucky, the representation might not be too bad when the basis is truncated to a small size.

The posterior weight distribution \(\boldsymbol{w} \mid \boldsymbol{y} \sim \mathcal{N}\left(\boldsymbol{\mu}_{\boldsymbol{w} \mid n}, \boldsymbol{\Sigma}_{\boldsymbol{w} \mid n}\right)\) is Gaussian with moments \[ \begin{aligned} \boldsymbol{\mu}_{\boldsymbol{w} \mid n} &=\left(\boldsymbol{\Phi}^{\top} \boldsymbol{\Phi}+\sigma^{2} \mathbf{I}\right)^{-1} \boldsymbol{\Phi}^{\top} \boldsymbol{y} \\ \boldsymbol{\Sigma}_{\boldsymbol{w} \mid n} &=\left(\boldsymbol{\Phi}^{\top} \boldsymbol{\Phi}+\sigma^{2} \mathbf{I}\right)^{-1} \sigma^{2} \end{aligned} \] where \(\boldsymbol{\Phi}=\boldsymbol{\phi}(\mathbf{X})\) is an \(n \times \ell\) feature matrix. We solve for the right-hand side at \(\mathcal{O}\left(\min \{\ell, n\}^{3}\right)\) cost by applying the Woodbury identity as needed. So far there is nothing unusual here; the cool bit is realising we can represent this update as an independent operation.

Figure 5: GP function updates from J. T. Wilson et al. (2021).

In the weight-space setting, the pathwise update given an initial weight vector \(\boldsymbol{w} \sim \mathcal{N}(\mathbf{0}, \mathbf{I})\) is \(\boldsymbol{w} \mid \boldsymbol{y} \stackrel{\mathrm{d}}{=} \boldsymbol{w}+\boldsymbol{\Phi}^{\top}\left(\boldsymbol{\Phi} \boldsymbol{\Phi}^{\top}+\sigma^{2} \mathbf{I}\right)^{-1}\left(\boldsymbol{y}-\boldsymbol{\Phi}^{\top} \boldsymbol{w}-\boldsymbol{\varepsilon}\right).\)

So if we had a nice weight-space representation for everything already we could go home at this point. However, for many models we are not given that; we might find natural bases for the prior and posterior are not the same (the posterior should not be stationary usually, for one thing).

The innovation in J. T. Wilson et al. (2020) is to make different choices of functional bases for prior and posterior updates. We can choose anything really, AFAICT. They suggest Fourier basis for prior and the canonical basis, i.e. the reproducing kernel basis \(k(\cdot,\vv{x})\) for the update. Then we have \[ \underbrace{(f \mid \boldsymbol{y})(\cdot)}_{\text {sparse posterior }} \stackrel{\mathrm{d}}{\approx} \underbrace{\sum_{i=1}^{\ell} w_{i} \phi_{i}(\cdot)}_{\text {weight-space prior}} +\underbrace{\sum_{j=1}^{m} v_{j} k\left(\cdot, \boldsymbol{x}_{j}\right)}_{\text {function-space update}} , \] where we have defined \(\boldsymbol{v}=\left(\mathbf{K}_{n, n}+\sigma^{2} \mathbf{I}\right)^{-1}\left(\boldsymbol{y}-\boldsymbol{\Phi}^{\top} \boldsymbol{w}- \boldsymbol{\varepsilon}\right) .\)

4 Sparse GP setting

I.e. using inducing variables.

Given \(q(\boldsymbol{u})\), we approximate posterior distributions as \[ p\left(\boldsymbol{f}_{*} \mid \boldsymbol{y}\right) \approx \int_{\mathbb{R}^{m}} p\left(\boldsymbol{f}_{*} \mid \boldsymbol{u}\right) q(\boldsymbol{u}) \mathrm{d} \boldsymbol{u} . \] If \(\boldsymbol{u} \sim \mathcal{N}\left(\boldsymbol{\mu}_{\boldsymbol{u}}, \boldsymbol{\Sigma}_{\boldsymbol{u}}\right)\), we compute this integral analytically to obtain a Gaussian distribution with mean and covariance \[ \begin{aligned} \boldsymbol{m}_{* \mid m} &=\mathbf{K}_{*, m} \mathbf{K}_{m, m}^{-1} \boldsymbol{\mu}_{m} \\ \mathbf{K}_{*, * \mid m} &=\mathbf{K}_{*, *}+\mathbf{K}_{*, m} \mathbf{K}_{m, m}^{-1}\left(\boldsymbol{\Sigma}_{\boldsymbol{u}}-\mathbf{K}_{m, m}\right) \mathbf{K}_{m, m}^{-1} \mathbf{K}_{m, *^{*}} \end{aligned} \]

\[ \begin{aligned} &\boldsymbol{f}_{*} \mid \boldsymbol{u} \stackrel{\mathrm{d}}{=} \boldsymbol{f}_{*}+\mathbf{K}_{*, m} \mathbf{K}_{m, m}^{-1}\left(\boldsymbol{u}-\boldsymbol{f}_{m}\right) \\ \end{aligned} \]

When sampling from sparse GPs we draw \(\boldsymbol{f}_{*}\) and \(\boldsymbol{f}_{m}\) together from the prior, and independently generate target values \(\boldsymbol{u} \sim q(\boldsymbol{u}) .\) \[ \underbrace{(f \mid \boldsymbol{u})(\cdot)}_{\text {sparse posterior }} \stackrel{\mathrm{d}}{\approx} \underbrace{\sum_{i=1}^{\ell} w_{i} \phi_{i}(\cdot)}_{\text {weight-space prior}} +\underbrace{\sum_{j=1}^{m} v_{j} k\left(\cdot, \boldsymbol{z}_{j}\right)}_{\text {function-space update}} , \] where we have defined \(\boldsymbol{v}=\mathbf{K}_{m, m}^{-1}\left(\boldsymbol{u}-\boldsymbol{\Phi}^{\top} \boldsymbol{w}\right) .\)

5 Matrix GPs

(Ritter et al. 2021 appendix D) reframes the Matheron update and generalises it to matrix-Gaussians. TBC.

Figure 6: The Matrix Gaussian pathwise update of Ritter et al. (2021).

6 Stationary moves

Thus far we have talked about moves updating a prior to a posterior; how about moves within a posterior?

We could try Langevin sampling, for example, or SG MCMC but these all seem to require inverting the covariance matrix so are not likely to be efficient in general. Can we do better?

7 Incoming

8 References

Abrahamsen. 1997. A Review of Gaussian Random Fields and Correlation Functions.”
Abt, and Welch. 1998. Fisher Information and Maximum-Likelihood Estimation of Covariance Parameters in Gaussian Stochastic Processes.” Canadian Journal of Statistics.
Altun, Smola, and Hofmann. 2004. Exponential Families for Conditional Random Fields.” In Proceedings of the 20th Conference on Uncertainty in Artificial Intelligence. UAI ’04.
Alvarado, and Stowell. 2018. Efficient Learning of Harmonic Priors for Pitch Detection in Polyphonic Music.” arXiv:1705.07104 [Cs, Stat].
Ambikasaran, Foreman-Mackey, Greengard, et al. 2015. Fast Direct Methods for Gaussian Processes.” arXiv:1403.6015 [Astro-Ph, Stat].
Bachoc, F., Gamboa, Loubes, et al. 2018. A Gaussian Process Regression Model for Distribution Inputs.” IEEE Transactions on Information Theory.
Bachoc, Francois, Suvorikova, Ginsbourger, et al. 2019. Gaussian Processes with Multidimensional Distribution Inputs via Optimal Transport and Hilbertian Embedding.” arXiv:1805.00753 [Stat].
Birgé, and Massart. 2006. Minimal Penalties for Gaussian Model Selection.” Probability Theory and Related Fields.
Bonilla, Chai, and Williams. 2007. Multi-Task Gaussian Process Prediction.” In Proceedings of the 20th International Conference on Neural Information Processing Systems. NIPS’07.
Bonilla, Krauth, and Dezfouli. 2019. Generic Inference in Latent Gaussian Process Models.” Journal of Machine Learning Research.
Borovitskiy, Terenin, Mostowsky, et al. 2020. Matérn Gaussian Processes on Riemannian Manifolds.” arXiv:2006.10160 [Cs, Stat].
Burt, Rasmussen, and Wilk. 2020. Convergence of Sparse Variational Inference in Gaussian Processes Regression.” Journal of Machine Learning Research.
Calandra, Peters, Rasmussen, et al. 2016. Manifold Gaussian Processes for Regression.” In 2016 International Joint Conference on Neural Networks (IJCNN).
Cressie. 1990. The Origins of Kriging.” Mathematical Geology.
———. 2015. Statistics for Spatial Data.
Cressie, and Wikle. 2011. Statistics for Spatio-Temporal Data. Wiley Series in Probability and Statistics 2.0.
Csató, and Opper. 2002. Sparse On-Line Gaussian Processes.” Neural Computation.
Csató, Opper, and Winther. 2001. TAP Gibbs Free Energy, Belief Propagation and Sparsity.” In Proceedings of the 14th International Conference on Neural Information Processing Systems: Natural and Synthetic. NIPS’01.
Cunningham, Shenoy, and Sahani. 2008. Fast Gaussian Process Methods for Point Process Intensity Estimation.” In Proceedings of the 25th International Conference on Machine Learning. ICML ’08.
Cutajar, Bonilla, Michiardi, et al. 2017. Random Feature Expansions for Deep Gaussian Processes.” In PMLR.
Dahl, and Bonilla. 2017. Scalable Gaussian Process Models for Solar Power Forecasting.” In Data Analytics for Renewable Energy Integration: Informing the Generation and Distribution of Renewable Energy. Lecture Notes in Computer Science.
Dahl, and Bonilla. 2019. Sparse Grouped Gaussian Processes for Solar Power Forecasting.” arXiv:1903.03986 [Cs, Stat].
Damianou, and Lawrence. 2013. Deep Gaussian Processes.” In Artificial Intelligence and Statistics.
Damianou, Titsias, and Lawrence. 2011. Variational Gaussian Process Dynamical Systems.” In Advances in Neural Information Processing Systems 24.
Dezfouli, and Bonilla. 2015. Scalable Inference for Gaussian Process Models with Black-Box Likelihoods.” In Advances in Neural Information Processing Systems 28. NIPS’15.
Domingos. 2020. Every Model Learned by Gradient Descent Is Approximately a Kernel Machine.” arXiv:2012.00152 [Cs, Stat].
Dubrule. 2018. Kriging, Splines, Conditional Simulation, Bayesian Inversion and Ensemble Kalman Filtering.” In Handbook of Mathematical Geosciences: Fifty Years of IAMG.
Dunlop, Girolami, Stuart, et al. 2018. How Deep Are Deep Gaussian Processes? Journal of Machine Learning Research.
Dutordoir, Hensman, van der Wilk, et al. 2021. Deep Neural Networks as Point Estimates for Deep Gaussian Processes.” In arXiv:2105.04504 [Cs, Stat].
Dutordoir, Saul, Ghahramani, et al. 2022. Neural Diffusion Processes.”
Duvenaud. 2014. Automatic Model Construction with Gaussian Processes.”
Duvenaud, Lloyd, Grosse, et al. 2013. Structure Discovery in Nonparametric Regression Through Compositional Kernel Search.” In Proceedings of the 30th International Conference on Machine Learning (ICML-13).
Ebden. 2015. Gaussian Processes: A Quick Introduction.” arXiv:1505.02965 [Math, Stat].
Eleftheriadis, Nicholson, Deisenroth, et al. 2017. Identification of Gaussian Process State Space Models.” In Advances in Neural Information Processing Systems 30.
Emery. 2007. Conditioning Simulations of Gaussian Random Fields by Ordinary Kriging.” Mathematical Geology.
Evgeniou, Micchelli, and Pontil. 2005. Learning Multiple Tasks with Kernel Methods.” Journal of Machine Learning Research.
Ferguson. 1973. A Bayesian Analysis of Some Nonparametric Problems.” The Annals of Statistics.
Finzi, Bondesan, and Welling. 2020. Probabilistic Numeric Convolutional Neural Networks.” arXiv:2010.10876 [Cs].
Föll, Haasdonk, Hanselmann, et al. 2017. Deep Recurrent Gaussian Process with Variational Sparse Spectrum Approximation.” arXiv:1711.00799 [Stat].
Frigola, Chen, and Rasmussen. 2014. Variational Gaussian Process State-Space Models.” In Advances in Neural Information Processing Systems 27.
Frigola, Lindsten, Schön, et al. 2013. Bayesian Inference and Learning in Gaussian Process State-Space Models with Particle MCMC.” In Advances in Neural Information Processing Systems 26.
Gal, and Ghahramani. 2015. Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning.” In Proceedings of the 33rd International Conference on Machine Learning (ICML-16).
Galliani, Dezfouli, Bonilla, et al. 2017. Gray-Box Inference for Structured Gaussian Process Models.” In Proceedings of the 20th International Conference on Artificial Intelligence and Statistics.
Gal, and van der Wilk. 2014. Variational Inference in Sparse Gaussian Process Regression and Latent Variable Models - a Gentle Tutorial.” arXiv:1402.1412 [Stat].
Gardner, Pleiss, Bindel, et al. 2018. GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration.” In Proceedings of the 32nd International Conference on Neural Information Processing Systems. NIPS’18.
Gardner, Pleiss, Wu, et al. 2018. Product Kernel Interpolation for Scalable Gaussian Processes.” arXiv:1802.08903 [Cs, Stat].
Garnelo, Rosenbaum, Maddison, et al. 2018. Conditional Neural Processes.” arXiv:1807.01613 [Cs, Stat].
Garnelo, Schwarz, Rosenbaum, et al. 2018. Neural Processes.”
Ghahramani. 2013. Bayesian Non-Parametrics and the Probabilistic Approach to Modelling.” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.
Gilboa, Saatçi, and Cunningham. 2015. Scaling Multidimensional Inference for Structured Gaussian Processes.” IEEE Transactions on Pattern Analysis and Machine Intelligence.
Girolami, and Rogers. 2005. Hierarchic Bayesian Models for Kernel Learning.” In Proceedings of the 22nd International Conference on Machine Learning - ICML ’05.
Gramacy. 2016. laGP: Large-Scale Spatial Modeling via Local Approximate Gaussian Processes in R.” Journal of Statistical Software.
Gramacy, and Apley. 2015. Local Gaussian Process Approximation for Large Computer Experiments.” Journal of Computational and Graphical Statistics.
Gratiet, Marelli, and Sudret. 2016. Metamodel-Based Sensitivity Analysis: Polynomial Chaos Expansions and Gaussian Processes.” In Handbook of Uncertainty Quantification.
Grosse, Salakhutdinov, Freeman, et al. 2012. Exploiting Compositionality to Explore a Large Space of Model Structures.” In Proceedings of the Conference on Uncertainty in Artificial Intelligence.
Hartikainen, and Särkkä. 2010. Kalman Filtering and Smoothing Solutions to Temporal Gaussian Process Regression Models.” In 2010 IEEE International Workshop on Machine Learning for Signal Processing.
Hensman, Fusi, and Lawrence. 2013. Gaussian Processes for Big Data.” In Proceedings of the Twenty-Ninth Conference on Uncertainty in Artificial Intelligence. UAI’13.
Huber. 2014. Recursive Gaussian Process: On-Line Regression and Learning.” Pattern Recognition Letters.
Huggins, Campbell, Kasprzak, et al. 2018. Scalable Gaussian Process Inference with Finite-Data Mean and Variance Guarantees.” arXiv:1806.10234 [Cs, Stat].
Jankowiak, Pleiss, and Gardner. 2020. Deep Sigma Point Processes.” In Conference on Uncertainty in Artificial Intelligence.
Jordan. 1999. Learning in Graphical Models.
Karvonen, and Särkkä. 2016. Approximate State-Space Gaussian Processes via Spectral Transformation.” In 2016 IEEE 26th International Workshop on Machine Learning for Signal Processing (MLSP).
Kasim, Watson-Parris, Deaconu, et al. 2020. Up to Two Billion Times Acceleration of Scientific Simulations with Deep Neural Architecture Search.” arXiv:2001.08055 [Physics, Stat].
Kingma, and Welling. 2014. Auto-Encoding Variational Bayes.” In ICLR 2014 Conference.
Kocijan, Girard, Banko, et al. 2005. Dynamic Systems Identification with Gaussian Processes.” Mathematical and Computer Modelling of Dynamical Systems.
Ko, and Fox. 2009. GP-BayesFilters: Bayesian Filtering Using Gaussian Process Prediction and Observation Models.” In Autonomous Robots.
Krauth, Bonilla, Cutajar, et al. 2016. AutoGP: Exploring the Capabilities and Limitations of Gaussian Process Models.” In UAI17.
Krige. 1951. A Statistical Approach to Some Basic Mine Valuation Problems on the Witwatersrand.” Journal of the Southern African Institute of Mining and Metallurgy.
Kroese, and Botev. 2013. Spatial Process Generation.” arXiv:1308.0399 [Stat].
Lawrence, Neil. 2005. Probabilistic Non-Linear Principal Component Analysis with Gaussian Process Latent Variable Models.” Journal of Machine Learning Research.
Lawrence, Neil, Seeger, and Herbrich. 2003. Fast Sparse Gaussian Process Methods: The Informative Vector Machine.” In Proceedings of the 16th Annual Conference on Neural Information Processing Systems.
Lawrence, Neil D., and Urtasun. 2009. Non-Linear Matrix Factorization with Gaussian Processes.” In Proceedings of the 26th Annual International Conference on Machine Learning. ICML ’09.
Lázaro-Gredilla, Quiñonero-Candela, Rasmussen, et al. 2010. Sparse Spectrum Gaussian Process Regression.” Journal of Machine Learning Research.
Lee, Bahri, Novak, et al. 2018. Deep Neural Networks as Gaussian Processes.” In ICLR.
Leibfried, Dutordoir, John, et al. 2022. A Tutorial on Sparse Gaussian Processes and Variational Inference.”
Lenk. 2003. Bayesian Semiparametric Density Estimation and Model Verification Using a Logistic–Gaussian Process.” Journal of Computational and Graphical Statistics.
Lindgren, Rue, and Lindström. 2011. An Explicit Link Between Gaussian Fields and Gaussian Markov Random Fields: The Stochastic Partial Differential Equation Approach.” Journal of the Royal Statistical Society: Series B (Statistical Methodology).
Liutkus, Badeau, and Richard. 2011. Gaussian Processes for Underdetermined Source Separation.” IEEE Transactions on Signal Processing.
Lloyd, Duvenaud, Grosse, et al. 2014. Automatic Construction and Natural-Language Description of Nonparametric Regression Models.” In Twenty-Eighth AAAI Conference on Artificial Intelligence.
Louizos, Shi, Schutte, et al. 2019. The Functional Neural Process.” In Advances in Neural Information Processing Systems.
Lu. 2022. A Rigorous Introduction to Linear Models.”
MacKay. 1998. Introduction to Gaussian Processes.” NATO ASI Series. Series F: Computer and System Sciences.
———. 2002. Gaussian Processes.” In Information Theory, Inference & Learning Algorithms.
Matheron. 1963a. Traité de Géostatistique Appliquée. 2. Le Krigeage.
———. 1963b. Principles of Geostatistics.” Economic Geology.
Matthews, van der Wilk, Nickson, et al. 2016. GPflow: A Gaussian Process Library Using TensorFlow.” arXiv:1610.08733 [Stat].
Mattos, Dai, Damianou, et al. 2016. Recurrent Gaussian Processes.” In Proceedings of ICLR.
Mattos, Dai, Damianou, et al. 2017. Deep Recurrent Gaussian Processes for Outlier-Robust System Identification.” Journal of Process Control, DYCOPS-CAB 2016,.
Micchelli, and Pontil. 2005a. Learning the Kernel Function via Regularization.” Journal of Machine Learning Research.
———. 2005b. On Learning Vector-Valued Functions.” Neural Computation.
Minh. 2022. Finite Sample Approximations of Exact and Entropic Wasserstein Distances Between Covariance Operators and Gaussian Processes.” SIAM/ASA Journal on Uncertainty Quantification.
Mohammadi, Challenor, and Goodfellow. 2021. Emulating Computationally Expensive Dynamical Simulators Using Gaussian Processes.” arXiv:2104.14987 [Stat].
Moreno-Muñoz, Artés-Rodríguez, and Álvarez. 2019. Continual Multi-Task Gaussian Processes.” arXiv:1911.00002 [Cs, Stat].
Nagarajan, Peters, and Nevat. 2018. Spatial Field Reconstruction of Non-Gaussian Random Fields: The Tukey G-and-H Random Process.” SSRN Electronic Journal.
Nickisch, Solin, and Grigorevskiy. 2018. State Space Gaussian Processes with Non-Gaussian Likelihood.” In International Conference on Machine Learning.
O’Hagan. 1978. Curve Fitting and Optimal Design for Prediction.” Journal of the Royal Statistical Society: Series B (Methodological).
Papaspiliopoulos, Pokern, Roberts, et al. 2012. Nonparametric Estimation of Diffusions: A Differential Equations Approach.” Biometrika.
Pinder, and Dodd. 2022. GPJax: A Gaussian Process Framework in JAX.” Journal of Open Source Software.
Pleiss, Gardner, Weinberger, et al. 2018. Constant-Time Predictive Distributions for Gaussian Processes.” In.
Pleiss, Jankowiak, Eriksson, et al. 2020. Fast Matrix Square Roots with Applications to Gaussian Processes and Bayesian Optimization.” Advances in Neural Information Processing Systems.
Qi, Abdel-Gawad, and Minka. 2010. Sparse-Posterior Gaussian Processes for General Likelihoods.” In Proceedings of the Twenty-Sixth Conference on Uncertainty in Artificial Intelligence. UAI’10.
Quiñonero-Candela, and Rasmussen. 2005. A Unifying View of Sparse Approximate Gaussian Process Regression.” Journal of Machine Learning Research.
Raissi, and Karniadakis. 2017. Machine Learning of Linear Differential Equations Using Gaussian Processes.” arXiv:1701.02440 [Cs, Math, Stat].
Rasmussen, and Williams. 2006. Gaussian Processes for Machine Learning. Adaptive Computation and Machine Learning.
Reece, and Roberts. 2010. An Introduction to Gaussian Processes for the Kalman Filter Expert.” In 2010 13th International Conference on Information Fusion.
Ritter, Kukla, Zhang, et al. 2021. Sparse Uncertainty Representation in Deep Learning with Inducing Weights.” arXiv:2105.14594 [Cs, Stat].
Riutort-Mayol, Bürkner, Andersen, et al. 2020. Practical Hilbert Space Approximate Bayesian Gaussian Processes for Probabilistic Programming.” arXiv:2004.11408 [Stat].
Rossi, Heinonen, Bonilla, et al. 2021. Sparse Gaussian Processes Revisited: Bayesian Approaches to Inducing-Variable Approximations.” In Proceedings of The 24th International Conference on Artificial Intelligence and Statistics.
Saatçi. 2012. Scalable inference for structured Gaussian process models.”
Saatçi, Turner, and Rasmussen. 2010. Gaussian Process Change Point Models.” In Proceedings of the 27th International Conference on International Conference on Machine Learning. ICML’10.
Saemundsson, Terenin, Hofmann, et al. 2020. Variational Integrator Networks for Physically Structured Embeddings.” arXiv:1910.09349 [Cs, Stat].
Salimbeni, and Deisenroth. 2017. Doubly Stochastic Variational Inference for Deep Gaussian Processes.” In Advances In Neural Information Processing Systems.
Salimbeni, Eleftheriadis, and Hensman. 2018. Natural Gradients in Practice: Non-Conjugate Variational Inference in Gaussian Process Models.” In International Conference on Artificial Intelligence and Statistics.
Särkkä. 2011. Linear Operators and Stochastic Partial Differential Equations in Gaussian Process Regression.” In Artificial Neural Networks and Machine Learning – ICANN 2011. Lecture Notes in Computer Science.
———. 2013. Bayesian Filtering and Smoothing. Institute of Mathematical Statistics Textbooks 3.
Särkkä, and Hartikainen. 2012. Infinite-Dimensional Kalman Filtering Approach to Spatio-Temporal Gaussian Process Regression.” In Artificial Intelligence and Statistics.
Särkkä, Solin, and Hartikainen. 2013. Spatiotemporal Learning via Infinite-Dimensional Bayesian Filtering and Smoothing: A Look at Gaussian Process Regression Through Kalman Filtering.” IEEE Signal Processing Magazine.
Schulam, and Saria. 2017. Reliable Decision Support Using Counterfactual Models.” In Proceedings of the 31st International Conference on Neural Information Processing Systems. NIPS’17.
Shah, Wilson, and Ghahramani. 2014. Student-t Processes as Alternatives to Gaussian Processes.” In Artificial Intelligence and Statistics.
Sidén. 2020. Scalable Bayesian Spatial Analysis with Gaussian Markov Random Fields. Linköping Studies in Statistics.
Smith, Alvarez, and Lawrence. 2018. Gaussian Process Regression for Binned Data.” arXiv:1809.02010 [Cs, Stat].
Snelson, and Ghahramani. 2005. Sparse Gaussian Processes Using Pseudo-Inputs.” In Advances in Neural Information Processing Systems.
Solin, and Särkkä. 2020. Hilbert Space Methods for Reduced-Rank Gaussian Process Regression.” Statistics and Computing.
Tait, and Damoulas. 2020. Variational Autoencoding of PDE Inverse Problems.” arXiv:2006.15641 [Cs, Stat].
Tang, Zhang, and Banerjee. 2019. On Identifiability and Consistency of the Nugget in Gaussian Spatial Process Models.” arXiv:1908.05726 [Math, Stat].
Titsias, Michalis K. 2009a. Variational Learning of Inducing Variables in Sparse Gaussian Processes.” In International Conference on Artificial Intelligence and Statistics.
———. 2009b. Variational Model Selection for Sparse Gaussian Process Regression: TEchical Supplement.”
Titsias, Michalis, and Lawrence. 2010. Bayesian Gaussian Process Latent Variable Model.” In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics.
Tokdar. 2007. Towards a Faster Implementation of Density Estimation With Logistic Gaussian Process Priors.” Journal of Computational and Graphical Statistics.
Turner, Ryan, Deisenroth, and Rasmussen. 2010. State-Space Inference and Learning with Gaussian Processes.” In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics.
Turner, Richard E., and Sahani. 2014. Time-Frequency Analysis as Probabilistic Inference.” IEEE Transactions on Signal Processing.
van der Wilk, Wilson, and Rasmussen. 2014. “Variational Inference for Latent Variable Modelling of Correlation Structure.” In NIPS 2014 Workshop on Advances in Variational Inference.
Vanhatalo, Riihimäki, Hartikainen, et al. 2013. GPstuff: Bayesian Modeling with Gaussian Processes.” Journal of Machine Learning Research.
———, et al. 2015. Bayesian Modeling with Gaussian Processes Using the GPstuff Toolbox.” arXiv:1206.5754 [Cs, Stat].
Walder, Christian, Kim, and Schölkopf. 2008. Sparse Multiscale Gaussian Process Regression.” In Proceedings of the 25th International Conference on Machine Learning. ICML ’08.
Walder, C., Schölkopf, and Chapelle. 2006. Implicit Surface Modelling with a Globally Regularised Basis of Compact Support.” Computer Graphics Forum.
Wang, Pleiss, Gardner, et al. 2019. Exact Gaussian Processes on a Million Data Points.” In Advances in Neural Information Processing Systems.
Wikle, Cressie, and Zammit-Mangion. 2019. Spatio-Temporal Statistics with R.
Wilkinson, Andersen, Reiss, et al. 2019. End-to-End Probabilistic Inference for Nonstationary Audio Analysis.” arXiv:1901.11436 [Cs, Eess, Stat].
Wilkinson, Särkkä, and Solin. 2021. Bayes-Newton Methods for Approximate Bayesian Inference with PSD Guarantees.”
Williams, Christopher, Klanke, Vijayakumar, et al. 2009. Multi-Task Gaussian Process Learning of Robot Inverse Dynamics.” In Advances in Neural Information Processing Systems 21.
Williams, Christopher KI, and Seeger. 2001. Using the Nyström Method to Speed Up Kernel Machines.” In Advances in Neural Information Processing Systems.
Wilson, Andrew Gordon, and Adams. 2013. Gaussian Process Kernels for Pattern Discovery and Extrapolation.” In International Conference on Machine Learning.
Wilson, James T, Borovitskiy, Terenin, et al. 2020. Efficiently Sampling Functions from Gaussian Process Posteriors.” In Proceedings of the 37th International Conference on Machine Learning.
Wilson, James T, Borovitskiy, Terenin, et al. 2021. Pathwise Conditioning of Gaussian Processes.” Journal of Machine Learning Research.
Wilson, Andrew Gordon, Dann, Lucas, et al. 2015. The Human Kernel.” arXiv:1510.07389 [Cs, Stat].
Wilson, Andrew Gordon, and Ghahramani. 2011. Generalised Wishart Processes.” In Proceedings of the Twenty-Seventh Conference on Uncertainty in Artificial Intelligence. UAI’11.
———. 2012. “Modelling Input Varying Correlations Between Multiple Responses.” In Machine Learning and Knowledge Discovery in Databases. Lecture Notes in Computer Science.
Wilson, Andrew Gordon, Knowles, and Ghahramani. 2012. Gaussian Process Regression Networks.” In Proceedings of the 29th International Coference on International Conference on Machine Learning. ICML’12.
Wilson, Andrew Gordon, and Nickisch. 2015. Kernel Interpolation for Scalable Structured Gaussian Processes (KISS-GP).” In Proceedings of the 32Nd International Conference on International Conference on Machine Learning - Volume 37. ICML’15.
Zhang, Walder, Bonilla, et al. 2020. Quantile Propagation for Wasserstein-Approximate Gaussian Processes.” In Proceedings of NeurIPS 2020.

Footnotes

  1. There are neat extensions to the non-Gaussian and sparse cases; that comes later.↩︎