Using observations to infer the values of some parameters corresponds to solving an 'inverse problem'. Practitioners usually seek the 'best solution' implied by the data, but observations should only be used to falsify possible solutions, not to deduce any particular solution.
This is a preview of subscription content, log in via an institution to check access.
References
Popper, K. The Logic of Scientific Discovery (Basic Books, 1959).
Stigler, M. S. The History of Statistics: The Measurement of Uncertainty Before 1900 (Belknap, Harvard, 1986).
Guide to the Expression of Uncertainty in Measurement (International Organization for Standardization, Switzerland, 1995).
Backus, G. Proc. Natl Acad. Sci. 65, 1–105 (1970); ibid. 65, 281–287 (1970); ibid. 67, 282–289 (1970).
Backus, G. & Gilbert, F. Philos. Trans. R. Soc. London 266, 123–192 (1970).
Tarantola, A. Inverse Problem Theory and Methods for Model Parameter Estimation (SIAM, 2005).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Tarantola, A. Popper, Bayes and the inverse problem. Nature Phys 2, 492–494 (2006). https://doi.org/10.1038/nphys375
Issue Date:
DOI: https://doi.org/10.1038/nphys375
- Springer Nature Limited
This article is cited by
-
Efficient Inverse Method for Structural Identification Considering Modeling and Response Uncertainties
Chinese Journal of Mechanical Engineering (2022)
-
Fibre optic distributed acoustic sensing of volcanic events
Nature Communications (2022)
-
Reducing uncertainty in seismic assessment of multiple masonry buildings based on monitored demolitions
Bulletin of Earthquake Engineering (2022)
-
Conditioning generative adversarial networks on nonlinear data for subsurface flow model calibration and uncertainty quantification
Computational Geosciences (2022)
-
Machine learning powered ellipsometry
Light: Science & Applications (2021)