Electronic Journal of Statistics

Poisson intensity estimation with reproducing kernels

Seth Flaxman, Yee Whye Teh, and Dino Sejdinovic

Full-text: Open access


Despite the fundamental nature of the inhomogeneous Poisson process in the theory and application of stochastic processes, and its attractive generalizations (e.g. Cox process), few tractable nonparametric modeling approaches of intensity functions exist, especially when observed points lie in a high-dimensional space. In this paper we develop a new, computationally tractable Reproducing Kernel Hilbert Space (RKHS) formulation for the inhomogeneous Poisson process. We model the square root of the intensity as an RKHS function. Whereas RKHS models used in supervised learning rely on the so-called representer theorem, the form of the inhomogeneous Poisson process likelihood means that the representer theorem does not apply. However, we prove that the representer theorem does hold in an appropriately transformed RKHS, guaranteeing that the optimization of the penalized likelihood can be cast as a tractable finite-dimensional problem. The resulting approach is simple to implement, and readily scales to high dimensions and large-scale datasets.

Article information

Electron. J. Statist. Volume 11, Number 2 (2017), 5081-5104.

Received: June 2017
First available in Project Euclid: 15 December 2017

Permanent link to this document

Digital Object Identifier

Primary: 62G05: Estimation 60G55: Point processes 46E22: Hilbert spaces with reproducing kernels (= [proper] functional Hilbert spaces, including de Branges-Rovnyak and other structured spaces) [See also 47B32]

Nonparametric statistics computational statistics spatial statistics intensity estimation reproducing kernel Hilbert space inhomogeneous Poisson processes

Creative Commons Attribution 4.0 International License.


Flaxman, Seth; Teh, Yee Whye; Sejdinovic, Dino. Poisson intensity estimation with reproducing kernels. Electron. J. Statist. 11 (2017), no. 2, 5081--5104. doi:10.1214/17-EJS1339SI. https://projecteuclid.org/euclid.ejs/1513306868

Export citation


  • [1] Adams, R. P., Murray, I. and MacKay, D. J. (2009). Tractable nonparametric Bayesian inference in Poisson processes with Gaussian process intensities. In, Proceedings of the 26th Annual International Conference on Machine Learning 9–16. ACM.
  • [2] Bach, F. (2015). On the Equivalence between Quadrature Rules and Random Features., arXiv:1502.06800.
  • [3] Baker, C. T. H. (1977)., The Numerical Treatment of Integral Equations. Monographs on Numerical Analysis Series. Oxford: Clarendon Press.
  • [4] Bartoszynski, R., Brown, B. W., McBride, C. M. and Thompson, J. R. (1981). Some nonparametric techniques for estimating the intensity function of a cancer related nonstationary Poisson process., The Annals of Statistics 1050–1060.
  • [5] Berlinet, A. and Thomas-Agnan, C. (2004)., Reproducing Kernel Hilbert Spaces in Probability and Statistics. Kluwer.
  • [6] Berman, M. and Diggle, P. (1989). Estimating weighted integrals of the second-order intensity of a spatial point process., Journal of the Royal Statistical Society. Series B (Methodological) 81–92.
  • [7] Brooks, M. M. and Marron, J. S. (1991). Asymptotic optimality of the least-squares cross-validation bandwidth for kernel estimates of intensity functions., Stochastic Processes and their Applications 38 157–165.
  • [8] Cressie, N. and Wikle, C. K. (2011)., Statistics for spatio-temporal data 465. Wiley.
  • [9] Csató, L., Opper, M. and Winther, O. (2001). TAP Gibbs Free Energy, Belief Propagation and Sparsity. In, Advances in Neural Information Processing Systems 657–663.
  • [10] Cunningham, J. P., Shenoy, K. V. and Sahani, M. (2008). Fast Gaussian process methods for point process intensity estimation. In, ICML 192–199. ACM.
  • [11] Diggle, P. (1985). A kernel method for smoothing point process data., Applied Statistics 138–147.
  • [12] Diggle, P. J., Moraga, P., Rowlingson, B., Taylor, B. M. et al. (2013). Spatial and spatio-temporal Log-Gaussian Cox processes: extending the geostatistical paradigm., Statistical Science 28 542–563.
  • [13] Fasshauer, G. E. and McCourt, M. J. (2012). Stable evaluation of Gaussian radial basis function interpolants., SIAM Journal on Scientific Computing 34 A737–A762.
  • [14] Flaxman, S., Teh, Y. W. and Sejdinovic, D. (2017). Poisson intensity estimation with reproducing kernels. In, Artificial Intelligence and Statistics (AISTATS) 270–279.
  • [15] Flaxman, S. R., Wilson, A. G., Neill, D. B., Nickisch, H. and Smola, A. J. (2015). Fast Kronecker inference in Gaussian processes with non-Gaussian likelihoods., International Conference on Machine Learning.
  • [16] Gilboa, E., Saatci, Y. and Cunningham, J. (2013). Scaling Multidimensional Inference for Structured Gaussian Processes., Pattern Analysis and Machine Intelligence, IEEE Transactions on PP 1–1.
  • [17] Illian, J. B., Sørbye, S. H., Rue, H. et al. (2012). A toolbox for fitting complex spatial point process models using integrated nested Laplace approximation (INLA)., The Annals of Applied Statistics 6 1499–1530.
  • [18] Jones, M. C. (1993). Simple boundary correction for kernel density estimation., Statistics and Computing 3 135–146.
  • [19] Kimeldorf, G. and Wahba, G. (1971). Some results on Tchebycheffian spline functions., Journal of Mathematical Analysis and Applications 33 82–95.
  • [20] Kingman, J. F. C. (1993)., Poisson processes. Oxford Studies in Probability 3. The Clarendon Press Oxford University Press, New York. Oxford Science Publications.
  • [21] Kom Samo, Y. L. and Roberts, S. (2015). Scalable Nonparametric Bayesian Inference on Point Processes with Gaussian Processes. In, ICML 2227–2236.
  • [22] Lloyd, C., Gunter, T., Osborne, M. and Roberts, S. (2015). Variational Inference for Gaussian Process Modulated Poisson Processes. In, ICML 1814–1822.
  • [23] McCullagh, P. and Møller, J. (2006). The permanental process., Advances in applied probability 873–888.
  • [24] Møller, J., Syversveen, A. R. and Waagepetersen, R. P. (1998). Log Gaussian Cox processes., Scandinavian Journal of Statistics 25 451–482.
  • [25] Muandet, K., Sriperumbudur, B. and Schölkopf, B. (2014). Kernel Mean Estimation via Spectral Filtering. In, Advances in Neural Information Processing Systems.
  • [26] Oates, C. J. and Girolami, M. A. (2016). Control Functionals for Quasi-Monte Carlo Integration. In, Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, AISTATS 2016, Cadiz, Spain, May 9–11, 2016 (A. Gretton and C. C. Robert, eds.). JMLR Workshop and Conference Proceedings 51 56–65. JMLR.org.
  • [27] Ramlau-Hansen, H. (1983). Smoothing Counting Process Intensities by Means of Kernel Functions., Ann. Statist. 11 453–466.
  • [28] Rasmussen, C. E. and Williams, C. K. (2006)., Gaussian processes for machine learning. MIT Press.
  • [29] Schölkopf, B. and Smola, A. J. (2002)., Learning with kernels: support vector machines, regularization, optimization and beyond. MIT Press.
  • [30] Silverman, B. W. (1982). On the Estimation of a Probability Density Function by the Maximum Penalized Likelihood Method., Ann. Statist. 10 795–810.
  • [31] Teh, Y. W. and Rao, V. (2011). Gaussian process modulated renewal processes. In, Advances in Neural Information Processing Systems 2474–2482.
  • [32] Wahba, G. (1990)., Spline models for observational data 59. Siam.
  • [33] Wied, D. and Weißbach, R. (2012). Consistency of the kernel density estimator: a survey., Statistical Papers 53 1–21.
  • [34] Williams, C. and Seeger, M. (2001). Using the Nyström method to speed up kernel machines. In, Advances in Neural Information Processing Systems 682–688.
  • [35] Wilson, A. G., Dann, C. and Nickisch, H. (2015). Thoughts on Massively Scalable Gaussian Processes., arXiv:1511.01870.
  • [36] Zhu, H., Williams, C. K., Rohwer, R. and Morciniec, M. (1997). Gaussian regression and optimal finite dimensional linear models., Technical report.