Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
126 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Correlational Lagrangian Schrödinger Bridge: Learning Dynamics with Population-Level Regularization (2402.10227v1)

Published 4 Feb 2024 in cs.LG and stat.ML

Abstract: Accurate modeling of system dynamics holds intriguing potential in broad scientific fields including cytodynamics and fluid mechanics. This task often presents significant challenges when (i) observations are limited to cross-sectional samples (where individual trajectories are inaccessible for learning), and moreover, (ii) the behaviors of individual particles are heterogeneous (especially in biological systems due to biodiversity). To address them, we introduce a novel framework dubbed correlational Lagrangian Schr\"odinger bridge (CLSB), aiming to seek for the evolution "bridging" among cross-sectional observations, while regularized for the minimal population "cost". In contrast to prior methods relying on \textit{individual}-level regularizers for all particles \textit{homogeneously} (e.g. restraining individual motions), CLSB operates at the population level admitting the heterogeneity nature, resulting in a more generalizable modeling in practice. To this end, our contributions include (1) a new class of population regularizers capturing the temporal variations in multivariate relations, with the tractable formulation derived, (2) three domain-informed instantiations based on genetic co-expression stability, and (3) an integration of population regularizers into data-driven generative models as constrained optimization, and a numerical solution, with further extension to conditional generative models. Empirically, we demonstrate the superiority of CLSB in single-cell sequencing data analyses such as simulating cell development over time and predicting cellular responses to drugs of varied doses.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (54)
  1. Akhiezer, N. I. The classical moment problem and some related questions in analysis. SIAM, 2020.
  2. An introduction to stem cells. The Journal of Pathology: A Journal of the Pathological Society of Great Britain and Ireland, 197(4):419–423, 2002.
  3. Barr, A. H. The einstein summation notation. An Introduction to Physically Based Modeling (Course Notes 19), pages E, 1:57, 1991.
  4. Learning single-cell perturbation responses using neural optimal transport. Nature Methods, 20(11):1759–1768, 2023.
  5. Spatially resolved, highly multiplexed rna profiling in single cells. Science, 348(6233):aaa6090, 2015.
  6. Neural ordinary differential equations. Advances in neural information processing systems, 31, 2018.
  7. Deep momentum multi-marginal schr\\\backslash\” odinger bridge. arXiv preprint arXiv:2303.01751, 2023.
  8. Diffusion schrödinger bridge with applications to score-based generative modeling. Advances in Neural Information Processing Systems, 34:17695–17709, 2021.
  9. Causal identification of single-cell experimental perturbation effects with cinema-ot. Nature Methods, pp.  1–11, 2023.
  10. How to train your neural ode: the world of jacobian and kinetic regularization. In International conference on machine learning, pp.  3154–3164. PMLR, 2020.
  11. Kinetics and mechanism. The Journal of Physical Chemistry, 65(2):384–384, 1961.
  12. Biodiversity: an introduction. John Wiley & Sons, 2013.
  13. Geometric interpretation of gene coexpression network analysis. PLoS computational biology, 4(8):e1000117, 2008.
  14. Keller, G. Embryonic stem cell differentiation: emergence of a new era in biology and medicine. Genes & development, 19(10):1129–1155, 2005.
  15. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
  16. Neural lagrangian schr\\\backslash\” odinger bridge. arXiv preprint arXiv:2204.04853, 2022.
  17. Stochastic systems: Estimation, identification, and adaptive control. SIAM, 2015.
  18. Fluid mechanics. Academic press, 2015.
  19. Rna velocity of single cells. Nature, 560(7719):494–498, 2018.
  20. Scalable gradients for stochastic differential equations. In International Conference on Artificial Intelligence and Statistics, pp.  3870–3882. PMLR, 2020.
  21. Physical kinetics. In Textbook of theoretical physics. 10. 1983.
  22. Deep generalized schrödinger bridge. Advances in Neural Information Processing Systems, 35:9374–9388, 2022.
  23. A comparison of batch effect removal methods for enhancement of prediction performance using maqc-ii microarray gene expression data. The pharmacogenomics journal, 10(4):278–291, 2010.
  24. Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. Cell, 161(5):1202–1214, 2015.
  25. Riemannian continuous normalizing flows. Advances in Neural Information Processing Systems, 33:2503–2515, 2020.
  26. String: a database of predicted functional associations between proteins. Nucleic acids research, 31(1):258–261, 2003.
  27. Mikami, T. Optimal transportation problem as stochastic mechanics. Selected Papers on Probability and Statistics, Amer. Math. Soc. Transl. Ser, 2(227):75–94, 2008.
  28. Visualizing structure and transitions in high-dimensional biological data. Nature biotechnology, 37(12):1482–1492, 2019.
  29. Action matching: Learning stochastic dynamics from samples. 2023.
  30. Ot-flow: Fast and accurate continuous normalizing flows via optimal transport. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pp.  9223–9232, 2021.
  31. The biogrid interaction database: 2019 update. Nucleic acids research, 47(D1):D529–D541, 2019.
  32. Parzen, E. Stochastic processes. SIAM, 1999.
  33. Assessing the utility of gene co-expression stability in combination with correlation in the analysis of protein-protein interaction networks. In BMC genomics, volume 12, pp.  1–11. BioMed Central, 2011.
  34. Getting the measure of biodiversity. Nature, 405(6783):212–219, 2000.
  35. Risken, H. Fokker-planck equation. Springer, 1996.
  36. Santambrogio, F. Optimal transport for applied mathematicians. Birkäuser, NY, 55(58-63):94, 2015.
  37. Optimal-transport analysis of single-cell gene expression identifies developmental trajectories in reprogramming. Cell, 176(4):928–943, 2019.
  38. A tutorial on gaussian process regression: Modelling, exploring, and exploiting functions. Journal of Mathematical Psychology, 85:1–16, 2018.
  39. The problem of moments, volume 1. American Mathematical Society (RI), 1950.
  40. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020.
  41. Spanos, A. Probability theory and statistical inference: Empirical modeling with observational data. Cambridge University Press, 2019.
  42. Temporal dynamics of protein complexes in ppi networks: a case study using yeast cell cycle dynamics. In BMC bioinformatics, volume 13, pp.  1–9. BioMed Central, 2012.
  43. Massively multiplex chemical transcriptomics at single-cell resolution. Science, 367(6473):45–51, 2020.
  44. A gene-coexpression network for global discovery of conserved genetic modules. science, 302(5643):249–255, 2003.
  45. Temple, G. Gauss’s theorem in general relativity. Proceedings of the Royal Society of London. Series A-Mathematical and Physical Sciences, 154(882):354–363, 1936.
  46. Hemopoietic stem cell differentiation. Biochimica et Biophysica Acta (BBA)-Reviews on Cancer, 605(4):431–459, 1980.
  47. Trajectorynet: A dynamic optimal transport network for modeling cellular dynamics. In International conference on machine learning, pp.  9526–9536. PMLR, 2020.
  48. Neural stochastic differential equations: Deep latent gaussian models in the diffusion limit. arXiv preprint arXiv:1905.09883, 2019.
  49. Graph attention networks. arXiv preprint arXiv:1710.10903, 2017.
  50. Villani, C. et al. Optimal transport: old and new, volume 338. Springer, 2009.
  51. Principal component analysis. Chemometrics and intelligent laboratory systems, 2(1-3):37–52, 1987.
  52. Scalable unbalanced optimal transport using generative adversarial networks. arXiv preprint arXiv:1810.11447, 2018.
  53. Stem cells: past, present, and future. Stem cell research & therapy, 10(1):1–22, 2019.
  54. A novel approach to remove the batch effect of single-cell data. Cell discovery, 5(1):46, 2019.

Summary

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

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