


Determinantal point processes (DPPs) are popular tools that supply useful information for repulsiveness. They provide coherent probabilistic models when negative correlations arise and also represent new algorithms for inference problems like sampling, marginalization and conditioning. Recently, DPPs have played an increasingly important role in machine learning and statistics, since they are used for diverse subset selection problems. In this paper we use -DPP, a conditional DPP that models only sets of cardinality , to sample a diverse subset of species from a large phylogenetic tree. The tree sampling task is important in many studies in modern bioinformatics. The results show a fast mixing sampler for -DPP, for which a polynomial bound on the mixing time is given. This approach is applied to a real-world dataset of species, and we observe that leaves joined by a higher subtree are more likely to appear.
Mots clés : Determinantal point process, Kernel, Markov chain, Metropolis-Hasting, Mixing time, Phylogenetic tree
@article{MSIA_2021__10_1_1_0, author = {Diala Wehbe and Nicolas Wicker and Baydaa Al-Ayoubi and Luc Moulinier}, title = {Fixed-Size Determinantal Point Processes Sampling For Species Phylogeny}, journal = {MathematicS In Action}, pages = {1--13}, publisher = {Soci\'et\'e de Math\'ematiques Appliqu\'ees et Industrielles}, volume = {10}, number = {1}, year = {2021}, doi = {10.5802/msia.13}, language = {en}, url = {https://msia.centre-mersenne.org/item/MSIA_2021__10_1_1_0/} }
Diala Wehbe; Nicolas Wicker; Baydaa Al-Ayoubi; Luc Moulinier. Fixed-Size Determinantal Point Processes Sampling For Species Phylogeny. MathematicS In Action, Tome 10 (2021) no. 1, pp. 1-13. doi : 10.5802/msia.13. https://msia.centre-mersenne.org/item/MSIA_2021__10_1_1_0/
[1] Monte Carlo Markov chain algorithms for sampling strongly Rayleigh distributions and determinantal point processes, In COLT (2016)
[2] Negative dependence and the geometry of polynomials, J. Am. Math. Soc., Volume 22 (2009), pp. 521-567 | Article | MR 2476782 | Zbl 1206.62096
[3] On selecting a maximum volume sub-matrix of a matrix and related problems, Theor. Comput. Sci. (2009), pp. 4801-4811 | Article | MR 2583677 | Zbl 1181.15002
[4] Spectra of Graphs, Academic Press Inc., 1980 | Zbl 0458.05042
[5] Efficient volume sampling for row/column subset selection, FOCS (2010) no. 1-3-4, pp. 329-338
[6] Random Walks and Electric Networks, Mathematical Association of America, 1984 | Zbl 0583.60065
[7] Statistical ensembles of complex, quaternion,and real matrices, J. Math. Phys., Volume 6 (1965), pp. 440-449 | Article | MR 173726 | Zbl 0127.39304
[8] Random walks on graphs., Stochastic Processes Appl., Volume 2 (1974), pp. 311-336 | Article | MR 397887 | Zbl 0296.60046
[9] Fast determinantal point process sampling with application to clustering, NIPS (2013), pp. 2319-2327
[10] Spectral algorithms, Foundations and Trends in Theoretical Computer Science, Volume 4 (2009), pp. 157-288 | Article | MR 2558901
[11] Diffusion kernels on graphs and other discrete structures, Proceedings of the 19th International Conference On Machine Learning (2002), pp. 315-322
[12] kDPPs: fixed size determinantal point processes, Proceedings of the 28th International Conference on Machine Learning, Omnipress, 2011, pp. 1193-1200
[13] Determinantal point processes for machine learning, Foundations and Trends in Machine Learning, Volume 5 (2012) no. 2-3, pp. 123-286 | Article | Zbl 1278.68240
[14] Random walks on graphs: A survey, Combinatorics, Volume 2 (1993), pp. 1-46
[15] The Coincidence approach to stochastic point processes, Adv. Appl. Probab., Volume 7 (1975) no. 1, pp. 83-122 | Article | MR 380979 | Zbl 0366.60081
[16] On the density of eigenvalues of a random matrix, Nuclear Physics, Volume 18 (1960), pp. 420-427 | Article | MR 112895 | Zbl 0107.35702
[17] Laplacian matrices of graphs: A survey, Linear Algebra Appl., Volume 197-198 (1994), pp. 143-176 | Article | MR 1275613 | Zbl 0802.05053
[18] UniProt Protein Knowledgebase, Methods Mol. Biol, Volume 1558 (2017), pp. 41-55 | Article