16 June 2018 in ACM Transactions on Mathematical Software
We experimentally study the fundamental problem of computing the volume of a convex polytope given as an intersection of linear halfspaces. We implement and evaluate randomized polynomial-time algorithms for accurately approximating the polytope’s volume in high dimensions (e.g., few hundreds) based onhit-and-run random walks. To carry out this efficiently, we experimentally correlate the effect of parameters, such as random walk length and number of sample points, with accuracy and runtime. Our method is based on Monte Carlo algorithms with guaranteed speed and provably high probability of success for arbitrarily high precision. We exploit the problem’s features in implementing a practical rounding procedure of polytopes, in computing only partial “generations” of random points, and in designing fast polytope boundary oracles. Our publicly available software is significantly faster than exact computation and more accurate than existing approximation methods. For illustration, volume approximations of Birkhoff polytopesB11,…,B15are computed, in dimensions up to 196, whereas exact methods have only computed volumes of up toB10.
8 September 2017 in Global Ecology and Biogeography
CTConstantinos Tsirogiannis
Constantinos Tsirogiannis
JSJens‐Christian Svenning
+1
1 August 2013 in International Journal of Computational Geometry & Applications
We design an algorithm to compute the Newton polytope of the resultant, known as resultant polytope, or its orthogonal projection along a given direction. The resultant is fundamental in algebraic elimination, optimization, and geometric modeling. Our algorithm exactly computes vertex- and halfspace-representations of the polytope using an oracle producing resultant vertices in a given direction, thus avoiding walking on the polytope whose dimension is α - n - 1, where the input consists of α points in ℤn. Our approach is output-sensitive as it makes one oracle call per vertex and per facet. It extends to any polytope whose oracle-based definition is advantageous, such as the secondary and discriminant polytopes. Our publicly available implementation uses the experimental CGAL package triangulation. Our method computes 5-, 6- and 7- dimensional polytopes with 35K, 23K and 500 vertices, respectively, within 2hrs, and the Newton polytopes of many important surface equations encountered in geometric modeling in < 1sec, whereas the corresponding secondary polytopes are intractable. It is faster than tropical geometry software up to dimension 5 or 6. Hashing determinantal predicates accelerates execution up to 100 times. One variant computes inner and outer approximations with, respectively, 90% and 105% of the true volume, up to 25 times faster.
4 September 2018 in Mathematical Programming Computation
15 June 2023 in ACM Transactions on Mathematical Software
We introduce Reflective Hamiltonian Monte Carlo (ReHMC), an HMC-based algorithm to sample from a log-concave distribution restricted to a convex body. The random walk is based on incorporating reflections to the Hamiltonian dynamics such that the support of the target density is the convex body. We develop an efficient open source implementation of ReHMC and perform an experimental study on various high-dimensional datasets. The experiments suggest that ReHMC outperforms Hit-and-Run and Coordinate-Hit-and-Run regarding the time it needs to produce an independent sample, introducing practical truncated sampling in thousands of dimensions.