×

Machine learning line bundle cohomology. (English) Zbl 1535.32024

Summary: We investigate different approaches to machine learning of line bundle cohomology on complex surfaces as well as on Calabi-Yau three-folds. Standard function learning based on simple fully connected networks with logistic sigmoids is reviewed and its main features and shortcomings are discussed. It has been observed recently that line bundle cohomology can be described by dividing the Picard lattice into certain regions in each of which the cohomology dimension is described by a polynomial formula. Based on this structure, we set up a network capable of identifying the regions and their associated polynomials, thereby effectively generating a conjecture for the correct cohomology formula. For complex surfaces, we also set up a network which learns certain rigid divisors which appear in a recently discovered master formula for cohomology dimensions.
© 2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

MSC:

32Q25 Calabi-Yau theory (complex-analytic aspects)
32S60 Stratifications; constructible sheaves; intersection cohomology (complex-analytic aspects)
32J17 Compact complex \(3\)-folds
14J30 \(3\)-folds
14J32 Calabi-Yau manifolds (algebro-geometric aspects)
68T07 Artificial neural networks and deep learning

References:

[1] Y. H.He, Deep‐Learning the Landscape, arXiv:1706.02714.
[2] D.Krefl, R. K.Seong, Phys. Rev. D2017, 96, 066014, arXiv:1706.03346.
[3] F.Ruehle, JHEP2017, 1708, 038, arXiv:1706.07024.
[4] J.Carifio, J.Halverson, D.Krioukov, B. D.Nelson, JHEP2017, 1709, 157, arXiv:1707.00655.
[5] Y. H.He, Phys. Lett. B2017, 774, 564.
[6] K.Bull, Y. H.He, V.Jejjala, C.Mishra, Phys. Lett. B2018, 785, 65, arXiv:1806.03121.
[7] H.Erbin, S.Krippendorf, GANs for generating EFT models, arXiv:1809.02612. · Zbl 1472.81155
[8] Y. H.He, The Calabi‐Yau Landscape: from Geometry, to Physics, to Machine‐Learning, arXiv:1812.02893. · Zbl 1492.14001
[9] A.Cole, G.Shiu, JHEP2019, 1903, 054, arXiv:1812.06960.
[10] J.Halverson, B.Nelson, F.Ruehle, Branes with Brains: Exploring String Vacua with Deep Reinforcement Learning, arXiv:1903.11616. · Zbl 1416.83125
[11] A.Constantin, Heterotic String Models on Smooth Calabi‐Yau Threefolds. PhD thesis, Oxford U., 2013. arXiv:1808.09993.
[12] E. I.Buchbinder, A.Constantin, A.Lukas, JHEP2014, 1403, 025, arXiv:1311.1941.
[13] A.Constantin, A.Lukas, Formulae for Line Bundle Cohomology on Calabi‐Yau Threefolds, arXiv:1808.09992.
[14] M.Larfors, R.Schneider, Line bundle cohomologies on CICYs with Picard number two, arXiv:1906.00392.
[15] D.Klaewer, L.Schlechter, Phys. Lett. B2019, 789, 438, arXiv:1809.02547.
[16] C.Brodie, A.Constantin, R.Deen, A.Lukas, Topological Formulae for Line Bundle Cohomology on Surfaces, arXiv:1906.08363. · Zbl 1471.32022
[17] C.Brodie, A.Constantin, R.Deen, A.Lukas, Index Formulae for Line Bundle Cohomology on Complex Surfaces, arXiv:1906.08769.
[18] R.Hartshorne, Algebraic Geometry, Springer Science & Business Media, 2013. · Zbl 0532.14001
[19] P.Griffiths, J.Harris, Principles of Algebraic Geometry, John Wiley & Sons, 2014.
[20] T.Hübsch, Calabi‐Yau Manifolds: A Bestiary for Physicists, World Scientific, 1994. · Zbl 0771.53002
[21] R.Blumenhagen, B.Jurke, T.Rahn, H.Roschy, J. Math. Phys.2010, 51, 103525, arXiv:1003.5217.
[22] L. B.Anderson, Y. H.He, A.Lukas, JHEP20070707, 049, arXiv:hep‐th/0702210.
[23] J.Gray, Y. H.He, A.Ilderton, A.Lukas, JHEP2007, 0707, 023, arXiv:hep‐th/0703249.
[24] L. B.Anderson, Y. H.He, A.Lukas, JHEP2008, 0807, 104, arXiv:0805.2875.
[25] Y. H.He, S. J.Lee, A.Lukas, JHEP2010, 1005, 071, arXiv:0911.0865.
[26] L. B.Anderson, J.Gray, Y. H.He, A.Lukas, JHEP2010, 1002, 054, arXiv:0911.1569.
[27] R.Blumenhagen, V.Braun, T. W.Grimm, T.Weigand, Nucl. Phys. B2009, 815, 1, arXiv:0811.2936.
[28] P.Candelas, A. M.Dale, C. A.Lutken, R.Schimmrigk, Nucl. Phys. B1988, 298, 493.
[29] P.Green, T.Hubsch, Commun. Math. Phys.1987, 109, 99. · Zbl 0611.53055
[30] D. P.Kingma, J.Ba, Adam: A Method for Stochastic Optimization, arXiv:1412.6980.
[31] G.Cybenko, Math. Control Signal Systems1989, 2, 303. · Zbl 0679.94019
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.