Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash 86 tok/s
Gemini 2.5 Pro 51 tok/s Pro
GPT-5 Medium 43 tok/s
GPT-5 High 37 tok/s Pro
GPT-4o 98 tok/s
GPT OSS 120B 466 tok/s Pro
Kimi K2 225 tok/s Pro
2000 character limit reached

Inferring potential landscapes: A Schrödinger bridge approach to Maximum Caliber (2403.01357v1)

Published 3 Mar 2024 in cond-mat.stat-mech

Abstract: Schr\"odinger bridges have emerged as an enabling framework for unveiling the stochastic dynamics of systems based on marginal observations at different points in time. The terminology "bridge'' refers to a probability law that suitably interpolates such marginals. The theory plays a pivotal role in a variety of contemporary developments in machine learning, stochastic control, thermodynamics, and biology, to name a few, impacting disciplines such as single-cell genomics, meteorology, and robotics. In this work, we extend Schr\"odinger's paradigm of bridges to account for integral constraints along paths, in a way akin to Maximum Caliber - a Maximum Entropy principle applied in a dynamic context. The Maximum Caliber principle has proven useful to infer the dynamics of complex systems e.g., that model gene circuits and protein folding. We unify these two problems via a maximum likelihood formulation to reconcile stochastic dynamics with ensemble-path data. A variety of data types can be encompassed, ranging from distribution moments to average currents along paths. The framework enables inference of time-varying potential landscapes that drive the process. The resulting forces can be interpreted as the optimal control that drives the system in a way that abides by specified integral constraints. Analogous results are presented in a discrete-time, discrete-space setting and specialized to steady-state dynamics. We finish by illustrating the practical applicability of the framework through paradigmatic examples, such as that of bit erasure or protein folding. In doing so, we highlight the strengths of the proposed framework, namely, the generality of the theory, the ease of computation, and the ability to interpret results in terms of system dynamics. This is in contrast to Maximum-Caliber problems where the focus is typically on updating a probability law on paths.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (56)
  1. E. Schrödinger, Über die Umkehrung der Naturgesetze, Sitzungsberichte der Preuss Akad. Wissen. Phys. Math. Klasse, Sonderausgabe IX, 144 (1931).
  2. E. Schrödinger, Sur la théorie relativiste de l’électron et l’interprétation de la mécanique quantique, in Annales de l’institut Henri Poincaré, Vol. 2(4) (Presses universitaires de France, 1932) pp. 269–310.
  3. R. Chetrite, P. Muratore-Ginanneschi, and K. Schwieger, E. schrödinger’s 1931 paper “on the reversal of the laws of nature”[“über die umkehrung der naturgesetze”, sitzungsberichte der preussischen akademie der wissenschaften, physikalisch-mathematische klasse, 8 n9 144–153], The European Physical Journal H 46, 1 (2021).
  4. A. Wakolbinger, Schrödinger bridges from 1931 to 1991, Contribuciones en probabilidad y estadística matemática 3, pp. 61-79  (1992).
  5. C. Léonard, A survey of the Schrödinger problem and some of its connections with optimal transport, arXiv preprint arXiv:1308.0215  (2013).
  6. A. Dembo and O. Zeitouni, Large deviations techniques and applications, Vol. 38 (Springer Science & Business Media, 2009).
  7. K. Elamvazhuthi and S. Berman, Mean-field models in swarm robotics: A survey, Bioinspiration & Biomimetics 15, 015001 (2019).
  8. E. T. Jaynes, The minimum entropy production principle, Annual Review of Physical Chemistry 31, 579 (1980).
  9. T. Firman, G. Balázsi, and K. Ghosh, Building predictive models of genetic circuits using the principle of maximum caliber, Biophysical journal 113, 2121 (2017).
  10. H. Wan, G. Zhou, and V. A. Voelz, A maximum-caliber approach to predicting perturbed folding kinetics due to mutations, Journal of chemical theory and computation 12, 5768 (2016).
  11. Y. Chen, T. T. Georgiou, and M. Pavon, Stochastic control liaisons: Richard Sinkhorn meets Gaspard Monge on a Schrödinger bridge, Siam Review 63, 249 (2021a).
  12. Y. Chen, T. T. Georgiou, and M. Pavon, On the relation between optimal transport and Schrödinger bridges: A stochastic control viewpoint, Journal of Optimization Theory and Applications 169, 671 (2016a).
  13. I. Karatzas and S. Shreve, Brownian motion and stochastic calculus, Vol. 113 (Springer Science & Business Media, 2012).
  14. R. Fortet, Résolution d’un système d’équations de m. Schrödinger, Journal de Mathématiques Pures et Appliquées 19, 83 (1940).
  15. M. Essid and M. Pavon, Traversing the Schrödinger bridge strait: Robert Fortet’s marvelous proof redux, Journal of Optimization Theory and Applications 181, 23 (2019).
  16. C. Léonard, From the Schrödinger problem to the Monge–Kantorovich problem, Journal of Functional Analysis 262, 1879 (2012).
  17. E. Carlen, Stochastic mechanics: a look back and a look ahead, Diffusion, quantum theory and radically elementary mathematics 47, 117 (2014).
  18. M. Cuturi, Sinkhorn distances: Lightspeed computation of optimal transport, Advances in neural information processing systems 26 (2013).
  19. Y. Chen, T. Georgiou, and M. Pavon, Entropic and displacement interpolation: a computational approach using the Hilbert metric, SIAM Journal on Applied Mathematics 76, 2375 (2016b).
  20. Y. Chen, T. T. Georgiou, and M. Pavon, Controlling uncertainty, IEEE Control Systems Magazine 41, 82 (2021b).
  21. K. Sekimoto, Stochastic energetics (Springer Berlin, 2010).
  22. U. Seifert, From stochastic thermodynamics to thermodynamic inference, Annual Review of Condensed Matter Physics 10, 171 (2019).
  23. E. Aurell, C. Mejía-Monasterio, and P. Muratore-Ginanneschi, Optimal protocols and optimal transport in stochastic thermodynamics, Physical review letters 106, 250601 (2011).
  24. S. Ciliberto, Experiments in stochastic thermodynamics: Short history and perspectives, Physical Review X 7, 021051 (2017).
  25. J. L. Doob, Conditional brownian motion and the boundary limits of harmonic functions, Bulletin de la Société Mathématique de France 85, 431 (1957).
  26. C. Léonard, Feynman-kac formula under a finite entropy condition, Probability Theory and Related Fields 184, 1029 (2022).
  27. C. Léonard, Stochastic derivatives and generalized h-transforms of Markov processes, arXiv preprint arXiv:1102.3172  (2011).
  28. R. Chetrite and H. Touchette, Variational and optimal control representations of conditioned and driven processes, Journal of Statistical Mechanics: Theory and Experiment 2015, P12001 (2015).
  29. Y. Chen, T. T. Georgiou, and M. Pavon, The most likely evolution of diffusing and vanishing particles: Schrödinger bridges with unbalanced marginals, SIAM Journal on Control and Optimization 60, 2016 (2022).
  30. Y. Chen, T. T. Georgiou, and A. Tannenbaum, Stochastic control and nonequilibrium thermodynamics: Fundamental limits, IEEE transactions on automatic control 65, 2979 (2019).
  31. P. D. Dixit and K. A. Dill, Building Markov state models using optimal transport theory, The Journal of Chemical Physics 150 (2019).
  32. C. Léonard, Some properties of path measures, Séminaire de Probabilités XLVI , 207 (2014).
  33. P. Bolhuis, Z. Brotzakis, and B. Keller, Force field optimization by imposing kinetic constraints with path reweighting, arXiv preprint arXiv:2207.04558  (2022).
  34. P. D. Dixit and K. A. Dill, Inferring microscopic kinetic rates from stationary state distributions, Journal of chemical theory and computation 10, 3002 (2014).
  35. P. D. Dixit and K. A. Dill, Caliber corrected Markov modeling (c2m2): Correcting equilibrium Markov models, Journal of chemical theory and computation 14, 1111 (2018).
  36. P. D. Dixit, Stationary properties of maximum-entropy random walks, Physical Review E 92, 042149 (2015).
  37. P. D. Dixit, Communication: Introducing prescribed biases in out-of-equilibrium Markov models, The Journal of Chemical Physics 148 (2018).
  38. T. Mora, S. Deny, and O. Marre, Dynamical criticality in the collective activity of a population of retinal neurons, Phys. Rev. Lett. 114, 078105 (2015).
  39. L. Peliti and S. Pigolotti, Stochastic Thermodynamics: An Introduction (Princeton University Press, 2021).
  40. M. Pavon and F. Ticozzi, Discrete-time classical and quantum Markovian evolutions: Maximum entropy problems on path space, Journal of Mathematical Physics 51 (2010).
  41. T. T. Georgiou and M. Pavon, Positive contraction mappings for classical and quantum Schrödinger systems, Journal of Mathematical Physics 56 (2015).
  42. R. A. Horn and C. R. Johnson, Matrix analysis (Cambridge university press, 2012).
  43. We follow this approach for illustrative purposes. In general, it is more efficient to iterate over λ𝜆\lambdaitalic_λ using the gradient descent method described in equation (27).
  44. R. Landauer, Information is physical, Physics Today 44, 23 (1991).
  45. A. Bérut, A. Petrosyan, and S. Ciliberto, Information and thermodynamics: experimental verification of landauer’s erasure principle, Journal of Statistical Mechanics: Theory and Experiment 2015, P06015 (2015).
  46. J.-E. Shea and C. L. Brooks III, From folding theories to folding proteins: a review and assessment of simulation studies of protein folding and unfolding, Annual review of physical chemistry 52, 499 (2001).
  47. S. Yang, J. N. Onuchic, and H. Levine, Effective stochastic dynamics on a protein folding energy landscape, The Journal of chemical physics 125 (2006).
  48. J. Bechhoefer, Hidden Markov models for stochastic thermodynamics, New Journal of Physics 17, 075003 (2015).
  49. C.-J. Tsai, B. Ma, and R. Nussinov, Folding and binding cascades: shifts in energy landscapes, Proceedings of the National Academy of Sciences 96, 9970 (1999).
  50. K.-i. Okazaki and S. Takada, Dynamic energy landscape view of coupled binding and protein conformational change: induced-fit versus population-shift mechanisms, Proceedings of the National Academy of Sciences 105, 11182 (2008).
  51. P. Śledź and A. Caflisch, Protein structure-based drug design: from docking to molecular dynamics, Current opinion in structural biology 48, 93 (2018).
  52. A. Dickson, Mapping the ligand binding landscape, Biophysical journal 115, 1707 (2018).
  53. H. Wang and G. Oster, Energy transduction in the f1 motor of atp synthase, Nature 396, 279 (1998).
  54. S. Mukherjee and A. Warshel, Electrostatic origin of the mechanochemical rotary mechanism and the catalytic dwell of f1-atpase, Proceedings of the National Academy of Sciences 108, 20550 (2011).
  55. A. B. Kolomeisky and M. E. Fisher, Molecular motors: a theorist’s perspective, Annu. Rev. Phys. Chem. 58, 675 (2007).
  56. S. Panettieri and R. V. Ulijn, Energy landscaping in supramolecular materials, Current Opinion in Structural Biology 51, 9 (2018).
Citations (1)
View on Semantic Scholar
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now