Lecture Notes on Inverse Theory

08 July 2021, Version 1
This content is an early or alternative research output and has not been peer-reviewed by Cambridge University Press at the time of posting.

Abstract

Inverse theory is the art of inferring properties of a physical system from observations. It is used to transform recordings of wavefields into estimates of medium properties in seismic tomography, medical imaging and nondestructive testing. It serves to constrain the distribution of density inside the Earth from gravity measurements, and to obtain probabilities of life on exoplanets based on telescope observations of electromagnetic spectra. The list of examples is endless. The primary goal of these lecture notes is to equip the reader with a solid theoretical background and with an extensive toolbox that can be used to solve real-world inverse problems. A central theme is the recognition that the best theory or the best method do not exist. A good solution strategy is application- and data-specific, and so these notes are intended to help with making a reasonable choice.

Keywords

Inverse Theory
Probability Theory
Optimisation
Monte Carlo Methods
Scientific Inference
Wave Propagation
Wave-based Imaging

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting and Discussion Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.
Comment number 1, Haipeng Li: Nov 15, 2023, 06:23

Hi Andreas, Thank you for this wonderful book. I'm commenting here to find out if there's any chance of getting some Swiss chocolate! While reading, I noticed a typo on page 195, section 9.1. It seems you intended to write, "This is in stark contrast to the 2n solutions of the forward modelling equations, needed to approximate the partial derivatives of χ by finite differencing, as in equation (9.1)." However, the text currently reads "n + 1 solution," which refers to the first-order finite difference case, not the one in equation (9.1). Best,

Response,
Andreas Fichtner :
Dec 05, 2023, 09:39

Dear Haipeng Li, you are a very careful reader! Thank you very much! In the interest of my CO2 footprint, I would prefer to not ship chocolate all around the globe, but rather hand it to you during a conference or during other travels. Where are you based?

Response,
Haipeng Li :
Mar 12, 2024, 05:26

Hi Prof. Andreas, Sorry for the very late response. It's glad to hear from you back! I'm from SEP at Stanford and my supervisor Prof. Biondo is visiting ETH :). Looking forward to meeting you at AGU this year. Best, Haipeng