Machine Learning Line Bundle Cohomology

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.

[1]  R. Schneider,et al.  Line Bundle Cohomologies on CICYs with Picard Number Two , 2019, Fortschritte der Physik.

[2]  Fabian Ruehle,et al.  Branes with brains: exploring string vacua with deep reinforcement learning , 2019, Journal of High Energy Physics.

[3]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[4]  D. Krefl,et al.  Machine Learning of Calabi-Yau Volumes : arXiv , 2017, 1706.03346.

[5]  Vishnu Jejjala,et al.  Machine learning CICY threefolds , 2018, Physics Letters B.

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  C. Lütken,et al.  Complete intersection Calabi-Yau manifolds , 1988 .

[8]  Yang-Hui He,et al.  Monad bundles in heterotic string compactifications , 2008, 0805.2875.

[9]  Sven Krippendorf,et al.  GANs for generating EFT models , 2018, Physics Letters B.

[10]  Heterotic Compactification, An Algorithmic Approach , 2007, hep-th/0702210.

[11]  Dmitri Krioukov,et al.  Machine learning in the string landscape , 2017, Journal of High Energy Physics.

[12]  T. Willmore.  Algebraic Geometry , 1973, Nature.

[13]  Yang-Hui He,et al.  The Calabi-Yau Landscape: from Geometry, to Physics, to Machine-Learning , 2018, 1812.02893.

[14]  Lorenz Schlechter,et al.  Machine learning line bundle cohomologies of hypersurfaces in toric varieties , 2018, Physics Letters B.

[15]  E. Witten,et al.  Calabi‐Yau Manifolds: A Bestiary for Physicists , 1992 .

[16]  Tristan Hübsck,et al.  Calabi-Yau manifolds , 1992 .

[17]  A. Constantin,et al.  The moduli space of heterotic line bundle models: a case study for the tetra-quadric , 2013, 1311.1941.

[18]  A. Constantin.  Heterotic string models on smooth Calabi-Yau threefolds , 2018, 1808.09993.

[19]  P. Green,et al.  Calabi-Yau manifolds as complete intersections in products of complex projective spaces , 1987 .

[20]  G. Shiu,et al.  Topological data analysis for the string landscape , 2018, Journal of High Energy Physics.

[21]  A. Constantin,et al.  Formulae for Line Bundle Cohomology on Calabi‐Yau Threefolds , 2018, Fortschritte der Physik.

[22]  Callum R. Brodie,et al.  Topological Formulae for the Zeroth Cohomology of Line Bundles on Surfaces , 2019 .

[23]  Yang-Hui He,et al.  Heterotic models from vector bundles on toric Calabi-Yau manifolds , 2009, 0911.0865.

[24]  Cohomology of Line Bundles , 2004 .

[25]  Yang-Hui He,et al.  Exploring positive monad bundles and a new heterotic standard model , 2009, 0911.1569.

[26]  A new method for finding vacua in string phenomenology , 2007, hep-th/0703249.

[27]  Yang-Hui He,et al.  Machine-learning the string landscape , 2017 .

[28]  R. Blumenhagen,et al.  Cohomology of line bundles: A computational algorithm , 2010, 1003.5217.

[29]  Yang-Hui He,et al.  Deep-Learning the Landscape , 2017, 1706.02714.

[30]  Thomas W. Grimm,et al.  GUTs in Type IIB Orientifold Compactifications , 2008, 0811.2936.

[31]  Fabian Ruehle.  Evolving neural networks with genetic algorithms to study the string landscape , 2017, 1706.07024.

[32]  Joe W. Harris,et al.  Principles of Algebraic Geometry , 1978 .