Abstract
The geometric problem of estimating an unknown compact convex set from evaluations of its support function arises in a range of scientific and engineering applications. Traditional approaches typically rely on estimators that minimize the error over all possible compact convex sets; in particular, these methods allow for limited incorporation of prior structural information about the underlying set and the resulting estimates become increasingly more complicated to describe as the number of measurements available grows. We address both of these shortcomings by describing a framework for estimating tractably specified convex sets from support function evaluations. Building on the literature in convex optimization, our approach is based on estimators that minimize the error over structured families of convex sets that are specified as linear images of concisely described sets—such as the simplex or the spectraplex—in a higher-dimensional space that is not much larger than the ambient space. Convex sets parametrized in this manner are significant from a computational perspective as one can optimize linear functionals over such sets efficiently; they serve a different purpose in the inferential context of the present paper, namely, that of incorporating regularization in the reconstruction while still offering considerable expressive power. We provide a geometric characterization of the asymptotic behavior of our estimators, and our analysis relies on the property that certain sets which admit semialgebraic descriptions are Vapnik–Chervonenkis classes. Our numerical experiments highlight the utility of our framework over previous approaches in settings in which the measurements available are noisy or small in number as well as those in which the underlying set to be reconstructed is non-polyhedral.
Similar content being viewed by others
Notes
We note that this is the case even though the estimator \({\hat{K}}^{\mathrm {LSE}}_n\) is consistent; in particular, consistency simply refers to the convergence as \(n \rightarrow \infty \) of \({\hat{K}}^{\mathrm {LSE}}_n\) to \(K^\star \) in a topological sense (e.g., in Hausdorff distance) and it does not provide any information about the facial structure of \({\hat{K}}^{\mathrm {LSE}}_n\) relative to that of \(K^\star \).
The Tammes problem is a special case of Thompson’s problem as well as Smale’s 7th problem [29].
References
Averkov, G.: Optimal size of linear matrix inequalities in semidefinite approaches to polynomial optimization. SIAM J. Appl. Algebra Geom. 3(1), 128–151 (2019)
Balázs, G.: Convex Regression: Theory, Practice, and Applications. PhD thesis, University of Alberta (2016)
Barvinok, A.: Approximations of convex bodies by polytopes and by projections of spectrahedra (2012). arXiv:1204.0471
Ben-Tal, A., Nemirovski, A.: On polyhedral approximations of the second-order cone. Math. Oper. Res. 26(2), 193–205 (2001)
Bronstein, E.M.: Approximation of convex sets by polytopes. J. Math. Sci. (N.Y.) 153(6), 727–762 (2008)
Cai, T.T., Guntuboyina, A., Wei, Y.: Adaptive estimation of planar convex sets. Ann. Stat. 46(3), 1018–1049 (2018)
Danzer, L.: Finite point-sets on \(S^2\) with minimum distance as large as possible. Discrete Math. 60, 3–66 (1986)
Fisher, N.I., Hall, P., Turlach, B.A., Watson, G.S.: On the estimation of a convex set from noisy data on its support function. J. Am. Stat. Assoc. 92(437), 84–91 (1997)
Gaillard, F.: Normal chest CT-lung window. Radiopaedia. https://radiopaedia.org/cases/normal-chest-ct-lung-window
Gardner, R.J., Kiderlen, M.: A new algorithm for 3D reconstruction from support functions. IEEE Trans. Pattern Anal. Mach. Intell. 31(3), 556–562 (2009)
Gardner, R.J., Kiderlen, M., Milanfar, P.: Convergence of algorithms for reconstructing convex bodies and directional measures. Ann. Stat. 34(3), 1331–1374 (2006)
Gouveia, J., Parrilo, P.A., Thomas, R.R.: Lifts of convex sets and cone factorizations. Math. Oper. Res. 38(2), 248–264 (2013)
Gregor, J., Rannou, F.R.: Three-dimensional support function estimation and application for projection magnetic resonance imaging. Int. J. Imaging Syst. Technol. 12(1), 43–50 (2002)
Guntuboyina, A.: Optimal rates of convergence for convex set estimation from support functions. Ann. Stat. 40(1), 385–411 (2012)
Hall, P., Turlach, B.A.: On the estimation of a convex set with corners. IEEE Trans. Pattern Anal. Mach. Intell. 21(3), 225–234 (1999)
Hannah, L.A., Dunson, D.B.: Multivariate convex regression with adaptive partitioning. J. Mach. Learn. Res. 14, 3261–3294 (2013)
Kosorok, M.R.: Introduction to Empirical Processes and Semiparametric Inference. Springer Series in Statistics. Springer, New York (2008)
Lele, A.S., Kulkarni, S.R., Willsky, A.S.: Convex-polygon estimation from support-line measurements and applications to target reconstruction from laser-radar data. J. Optical Soc. Amer. A 9(10), 1693–1714 (1992)
Lloyd, S.P.: Least squares quantization in PCM. IEEE Trans. Inf. Theory 28(2), 129–137 (1982)
Magnani, A., Boyd, S.P.: Convex piecewise-linear fitting. Optim. Eng. 10(1), 1–17 (2009)
Nesterov, Yu., Nemirovskii, A.: Interior-Point Polynomial Algorithms in Convex Programming. SIAM Studies in Applied Mathematics, vol. 13. Society for Industrial and Applied Mathematics, Philadelphia (1994)
Pollard, D.: Strong consistency of \(k\)-means clustering. Ann. Stat. 9(1), 135–140 (1981)
Pollard, D.: Convergence of Stochastic Processes. Springer Series in Statistics. Springer, New York (1984)
Prince, J.L., Willsky, A.S.: Reconstructing convex sets from support line measurements. IEEE Trans. Pattern Anal. Mach. Intell. 12(4), 377–389 (1990)
Prince, J.L., Willsky, A.S.: Convex set reconstruction using prior shape information. CVGIP: Graph. Models Image Process. 53(5), 413–427 (1991)
Saunderson, J.: Limitations on the expressive power of convex cones without long chains of faces. SIAM J. Optim. 30(1), 1033–1047 (2020)
Schneider, R.: Convex Bodies: The Brunn–Minkowski Theory. Encyclopedia of Mathematics and Its Applications, vol. 151. Cambridge University Press, Cambridge (2014)
Schütte, K., van der Waerden, B.L.: Auf welcher Kugel haben \(5\), \(6\), \(7\), \(8\) oder \(9\) Punkte mit Mindestabstand Eins Platz? Math. Ann. 123, 96–124 (1951)
Smale, S.: Mathematical problems for the next century. Math. Intell. 20(2), 7–15 (1998)
Stark, H., Peng, H.: Shape estimation in computer tomography from minimal data. J. Opt. Soc. Am. A 5(3), 331–343 (1988)
Stengle, G., Yukich, J.E.: Some new Vapnik–Chervonenkis classes. Ann. Stat. 17(4), 1441–1446 (1989)
Tammes, P.M.L.: On the Origin of Number and Arrangement of the Places of Exit on the Surface of Pollen-Grains. Recueil des Travaux Botaniques Néerlandais, vol. 27. Koninklijke Nederlandse Botanische Vereniging (1930)
Vapnik, V.N., Chervonenkis, A.Ya.: On the uniform convergence of relative frequencies of events to their probabilities. Proc. USSR Acad. Sci. 181(4), 781–783 (1968). (in Russian)
Yannakakis, M.: Expressing combinatorial optimization problems by linear programs. J. Comput. Syst. Sci. 43(3), 441–466 (1991)
Author information
Authors and Affiliations
Corresponding author
Additional information
Editor in Charge: Kenneth Clarkson
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The authors were supported in part by NSF grants CCF-1350590 and CCF-1637598, by Air Force Office of Scientific Research Grant FA9550-16-1-0210, by a Sloan research fellowship, and an A*STAR (Agency for Science, Technology and Research, Singapore) fellowship.
Rights and permissions
About this article
Cite this article
Soh, Y.S., Chandrasekaran, V. Fitting Tractable Convex Sets to Support Function Evaluations. Discrete Comput Geom 66, 510–551 (2021). https://doi.org/10.1007/s00454-020-00258-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00454-020-00258-0