Skip to main content

Disentangling a deep learned volume formula

A preprint version of the article is available at arXiv.

Abstract

We present a simple phenomenological formula which approximates the hyperbolic volume of a knot using only a single evaluation of its Jones polynomial at a root of unity. The average error is just 2.86% on the first 1.7 million knots, which represents a large improvement over previous formulas of this kind. To find the approximation formula, we use layer-wise relevance propagation to reverse engineer a black box neural network which achieves a similar average error for the same approximation task when trained on 10% of the total dataset. The particular roots of unity which appear in our analysis cannot be written as e2πi/(k+2) with integer k; therefore, the relevant Jones polynomial evaluations are not given by unknot-normalized expectation values of Wilson loop operators in conventional SU(2) Chern-Simons theory with level k. Instead, they correspond to an analytic continuation of such expectation values to fractional level. We briefly review the continuation procedure and comment on the presence of certain Lefschetz thimbles, to which our approximation formula is sensitive, in the analytically continued Chern-Simons integration cycle.

References

  1. L. J. Dixon, Some worldsheet properties of superstring compactifications, on orbifolds and otherwise, in Summer Workshop in High-energy Physics and Cosmology, (1987) PUPT-1074.

  2. W. Lerche, C. Vafa and N. P. Warner, Chiral Rings in N = 2 Superconformal Theories, Nucl. Phys. B 324 (1989) 427 [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  3. P. Candelas, M. Lynker and R. Schimmrigk, Calabi-Yau Manifolds in Weighted P(4), Nucl. Phys. B 341 (1990) 383 [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  4. B. R. Greene and M. R. Plesser, Duality in Calabi-Yau Moduli Space, Nucl. Phys. B 338 (1990) 15 [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  5. P. Candelas, X. C. De La Ossa, P. S. Green and L. Parkes, A Pair of Calabi-Yau manifolds as an exactly soluble superconformal theory, Nucl. Phys. B 359 (1991) 21 [AMS/IP Stud. Adv. Math. 9 (1998) 31] [INSPIRE].

  6. A. Givental, A mirror theorem for toric complete intersections, in Topological field theory, primitive forms and related topics, Springer (1998), pp. 141–175.

  7. B. H. Lian, K. F. Liu and S.-T. Yau, Mirror principle. 1., Asian J. Math. 1 (1997) 729 [INSPIRE].

  8. V. F. R. Jones, Hecke algebra representations of braid groups and link polynomials, Annals Math. 126 (1987) 335 [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  9. W. Thurston, The Geometry and Topology of 3-Manifolds, Lecture Notes, Princeton University, New Jersey U.S.A. (1978).

    Google Scholar 

  10. E. Witten, Quantum Field Theory and the Jones Polynomial, Commun. Math. Phys. 121 (1989) 351 [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  11. R. M. Kashaev, The Hyperbolic volume of knots from quantum dilogarithm, Lett. Math. Phys. 39 (1997) 269 [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  12. H. Murakami and J. Murakami, The colored Jones polynomials and the simplicial volume of a knot, Acta Math. 186 (2001) 85 [math/9905075].

  13. S. Gukov, Three-dimensional quantum gravity, Chern-Simons theory, and the A polynomial, Commun. Math. Phys. 255 (2005) 577 [hep-th/0306165] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  14. N. Dunfield, An interesting relationship between the Jones polynomial and hyperbolic volume, (2000) https://faculty.math.illinois.edu/∼nmd/preprints/misc/dylan/index.html.

  15. S. Garoufalidis and Y. Lan, Experimental evidence for the volume conjecture for the simplest hyperbolic non-2-bridge knot, Algebr. Geom. Topol. 5 (2005) 379 [math/0412331].

  16. V. Jejjala, A. Kar and O. Parrikar, Deep Learning the Hyperbolic Volume of a Knot, Phys. Lett. B 799 (2019) 135033 [arXiv:1902.05547] [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  17. L. G. Valiant, A theory of the learnable, Commun. ACM 27 (1984) 1134.

    MATH  Article  Google Scholar 

  18. E. Witten, Analytic Continuation Of Chern-Simons Theory, AMS/IP Stud. Adv. Math. 50 (2011) 347 [arXiv:1001.2933] [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  19. W. P. Thurston, Three-Dimensional Geometry and Topology, Princeton University Press, (1997).

    MATH  Book  Google Scholar 

  20. V. Balasubramanian, J. R. Fliss, R. G. Leigh and O. Parrikar, Multi-Boundary Entanglement in Chern-Simons Theory and Link Invariants, JHEP 04 (2017) 061 [arXiv:1611.05460] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  21. V. Balasubramanian, M. DeCross, J. Fliss, A. Kar, R. G. Leigh and O. Parrikar, Entanglement Entropy and the Colored Jones Polynomial, JHEP 05 (2018) 038 [arXiv:1801.01131] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  22. T. Dimofte and S. Gukov, Quantum Field Theory and the Volume Conjecture, Contemp. Math. 541 (2011) 41 [arXiv:1003.4808] [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  23. Y. LeCun, Y. Bengio and G. Hinton, Deep learning, Nature 521 (2015) 436.

    Article  ADS  Google Scholar 

  24. M. C. Hughes, A neural network approach to predicting and computing knot invariants, (2016) arXiv:1610.05744.

  25. S. Gukov, J. Halverson, F. Ruehle and P. Sułkowski, Learning to Unknot, Mach. Learn. Sci. Tech. 2 (2021) 025035 [arXiv:2010.16263] [INSPIRE].

    Article  Google Scholar 

  26. G. Cybenko, Approximation by superpositions of a sigmoidal function, Math. Control Signals Syst. 2 (1989) 303.

    MathSciNet  MATH  Article  Google Scholar 

  27. K. Hornik, Approximation capabilities of multilayer feedforward networks, Neural Networks 4 (1991) 251.

    Article  Google Scholar 

  28. J. Hoste, M. Thistlethwaite and J. Weeks, The first 1,701,936 knots, Math. Intelligencer 20 (1998) 33.

    MathSciNet  MATH  Article  Google Scholar 

  29. The knot atlas: The take home database, www.katlas.org/wiki/The_Take_Home_Database.

  30. M. Culler, N. M. Dunfield, M. Goerner and J. R. Weeks, SnapPy, a computer program for studying the geometry and topology of 3-manifolds, http://snappy.computop.org.

  31. S. Bach, A. Binder, G. Montavon, F. Klauschen, K.-R. Müller and W. Samek, On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation, PloS one 10 (2015) e0130140.

    Article  Google Scholar 

  32. G. Montavon, A. Binder, S. Lapuschkin, W. Samek and K.-R. Müller, Layer-wise relevance propagation: An overview, in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, Springer (2019) [DOI].

  33. O. Dasbach and X.-S. Lin, A volumish theorem for the jones polynomial of alternating knots, Pacific J. Math. 231 (2007) 279.

    MathSciNet  MATH  Article  Google Scholar 

  34. S.-M. Udrescu and M. Tegmark, AI Feynman: a Physics-Inspired Method for Symbolic Regression, Sci. Adv. 6 (2020) eaay2631 [arXiv:1905.11481] [INSPIRE].

  35. S. Blücher, L. Kades, J. M. Pawlowski, N. Strodthoff and J. M. Urban, Towards novel insights in lattice field theory with explainable machine learning, Phys. Rev. D 101 (2020) 094507 [arXiv:2003.01504] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  36. J. S. Cotler et al., Black Holes and Random Matrices, JHEP 05 (2017) 118 [Erratum ibid. 09 (2018) 002] [arXiv:1611.04650] [INSPIRE].

  37. G. W. Moore, Introduction to Chern-Simons theories, lecture notes for a series of talks at the 2019 TASI school, (2019) http://www.physics.rutgers.edu/∼gmoore/TASI-ChernSimons-StudentNotes.pdf.

  38. M. Khovanov, Patterns in knot cohomology, I, Exper. Math. 12 (2003) 365.

    MathSciNet  MATH  Article  Google Scholar 

  39. K. Habiro, On the colored jones polynomials of some simple links, Sūrikaisekikenkyūsho Kōkyūroku, Kyoto University, Kyoto (2000).

    MATH  Google Scholar 

  40. D. Bar-Natan, The symbol font dbnsymb, (2020) https://www.math.toronto.edu/∼drorbn/projects/dbnsymb/dbnsymbman.html.

  41. M. Abadi et al., TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems, arXiv:1603.04467 [INSPIRE].

  42. Y.-H. He, Deep-Learning the Landscape, arXiv:1706.02714 [INSPIRE].

  43. F. Ruehle, Evolving neural networks with genetic algorithms to study the String Landscape, JHEP 08 (2017) 038 [arXiv:1706.07024] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  44. D. Krefl and R.-K. Seong, Machine Learning of Calabi-Yau Volumes, Phys. Rev. D 96 (2017) 066014 [arXiv:1706.03346] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  45. J. Carifio, J. Halverson, D. Krioukov and B. D. Nelson, Machine Learning in the String Landscape, JHEP 09 (2017) 157 [arXiv:1707.00655] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  46. K. Bull, Y.-H. He, V. Jejjala and C. Mishra, Machine Learning CICY Threefolds, Phys. Lett. B 785 (2018) 65 [arXiv:1806.03121] [INSPIRE].

    Article  ADS  Google Scholar 

  47. K. Bull, Y.-H. He, V. Jejjala and C. Mishra, Getting CICY High, Phys. Lett. B 795 (2019) 700 [arXiv:1903.03113] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  48. Y.-H. He and S.-T. Yau, Graph Laplacians, Riemannian Manifolds and their Machine-Learning, arXiv:2006.16619 [INSPIRE].

  49. Y.-H. He and A. Lukas, Machine Learning Calabi-Yau Four-folds, Phys. Lett. B 815 (2021) 136139 [arXiv:2009.02544] [INSPIRE].

    MathSciNet  Article  Google Scholar 

  50. J. Halverson, B. Nelson and F. Ruehle, Branes with Brains: Exploring String Vacua with Deep Reinforcement Learning, JHEP 06 (2019) 003 [arXiv:1903.11616] [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  51. J. Halverson, M. Plesser, F. Ruehle and J. Tian, Kähler Moduli Stabilization and the Propagation of Decidability, Phys. Rev. D 101 (2020) 046010 [arXiv:1911.07835] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  52. J. Halverson and C. Long, Statistical Predictions in String Theory and Deep Generative Models, Fortsch. Phys. 68 (2020) 2000005 [arXiv:2001.00555] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  53. M. Bies, M. Cvetič, R. Donagi, L. Lin, M. Liu and F. Ruehle, Machine Learning and Algebraic Approaches towards Complete Matter Spectra in 4d F-theory, JHEP 01 (2021) 196 [arXiv:2007.00009] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  54. F. Ruehle, Data science applications to string theory, Phys. Rept. 839 (2020) 1 [INSPIRE].

    MathSciNet  MATH  Article  ADS  Google Scholar 

  55. C. R. Brodie, A. Constantin, R. Deen and A. Lukas, Machine Learning Line Bundle Cohomology, Fortsch. Phys. 68 (2020) 1900087 [arXiv:1906.08730] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  56. C. R. Brodie, A. Constantin, R. Deen and A. Lukas, Topological Formulae for the Zeroth Cohomology of Line Bundles on Surfaces, arXiv:1906.08363 [INSPIRE].

  57. C. R. Brodie, A. Constantin, R. Deen and A. Lukas, Index Formulae for Line Bundle Cohomology on Complex Surfaces, Fortsch. Phys. 68 (2020) 1900086 [arXiv:1906.08769] [INSPIRE].

    MathSciNet  Article  ADS  Google Scholar 

  58. M. Khovanov, A categorification of the Jones polynomial, Duke Math. J. 101 (2000) 359 [math/9908171] [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  59. D. Bar-Natan, On khovanov’s categorification of the jones polynomial, Algebr. Geom. Topol. 2 2 (2002) 337 [math/0201043].

  60. L. H. Kauffman, State models and the jones polynomial, Topology 26 (1987) 395.

    MathSciNet  MATH  Article  Google Scholar 

  61. K. P. Walsh, Patterns and Stability in the Coefficients of the Colored Jones Polynomial, Ph.D. Thesis, University of California, San Diego (2014) https://roger.ucsd.edu:443/record=b8201173∼S9.

  62. K. P. Walsh, Higher order stability in the coefficients of the colored jones polynomial, J. Knot Theor. Ramifications 27 (2018) 1840010 [arXiv:1603.06957].

    MathSciNet  MATH  Article  Google Scholar 

  63. T. T. Lê, The colored jones polynomial and the aj conjecture, (2014) http://people.math.gatech.edu/∼letu/Papers/Lectures_Luminy_2014_new.pdf.

  64. V. F. Jones, The jones polynomial for dummies, (2014) math.berkeley.edu/∼vfr/jonesakl.pdf.

  65. V. F. Jones, The jones polynomial, (2005) https://math.berkeley.edu/∼vfr/jones.pdf.

  66. H. Murakami, J. Murakami, M. Okamoto, T. Takata and Y. Yokota, Kashaev’s conjecture and the Chern-Simons invariants of knots and links, Exper. Math. 11 (2002) 427.]

  67. L. van der Maaten and G. Hinton, Visualizing data using t-SNE, J. Mach. Learn. Res. 9 (2008) 2579.

    MATH  Google Scholar 

  68. F. Pedregosa et al., Scikit-learn: Machine learning in Python, J. Mach. Learn. Res. 12 (2011) 2825.

    MathSciNet  MATH  Google Scholar 

  69. P. Fréyd, D. Yetter, J. Hoste, W. B. R. Lickorish, K. Millett and A. Ocneanu, A new polynomial invariant of knots and links, Bull. Am. Math. Soc. 12 (1985) 239 [INSPIRE].

    MathSciNet  MATH  Article  Google Scholar 

  70. J. H. Przytycki and P. Traczyk, Conway algebras and skein equivalence of links, Proc. Am. Math. Soc. 100 (1987) 744.

    MathSciNet  MATH  Article  Google Scholar 

  71. J. W. Alexander, Topological invariants of knots and links, Trans. Am. Math. Soc. 30 (1928) 275.

    MathSciNet  MATH  Article  Google Scholar 

  72. M. Cranmer et al., Discovering Symbolic Models from Deep Learning with Inductive Biases, arXiv:2006.11287 [INSPIRE].

  73. M. Cranmer, Pysr: Fast & parallelized symbolic regression in python/julia, September 2020, https://doi.org/10.5281/zenodo.4052869.

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Jessica Craven.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

ArXiv ePrint: 2012.03955

Rights and permissions

Open Access . This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Craven, J., Jejjala, V. & Kar, A. Disentangling a deep learned volume formula. J. High Energ. Phys. 2021, 40 (2021). https://doi.org/10.1007/JHEP06(2021)040

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/JHEP06(2021)040

Keywords

  • Chern-Simons Theories
  • Topological Field Theories
  • Wilson
  • ’t Hooft and Polyakov loops