Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Inference from Inadequate and Inaccurate Data, III.

G Backus1

  • 1UNIVERSITY OF CALIFORNIA, LA JOLLA, CALIF. 92037.

Proceedings of the National Academy of Sciences of the United States of America
|September 1, 1970
PubMed
Summary

This study estimates unknown properties of physical objects using limited data. It introduces a method for estimating properties by bounding the Hilbert norm of projected models, enhancing predictive accuracy in complex systems.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Consumer acceptance of minced meat patties from boars in four European countries.

Meat science·2017
Same author

The influence of different glucose concentrations in haemodialysis solutions on metabolism and blood pressure stability in diabetic patients.

The International journal of artificial organs·2002
Same author

A simplified method for adequate hemodialysis.

Blood purification·2001
Same author

Long-term experience with an ultrapure individual dialysis fluid with a batch type machine.

Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association·1998
Same author

Isotropic probability measures in infinite-dimensional spaces.

Proceedings of the National Academy of Sciences of the United States of America·1987
Same author

Inference from Inadequate and Inaccurate Data, II.

Proceedings of the National Academy of Sciences of the United States of America·1970

Area of Science:

  • Physics
  • Statistics
  • Machine Learning

Background:

  • Estimating properties of complex physical objects from limited measurements is a significant challenge in scientific research.
  • Existing methods often struggle with high-dimensional data and uncertainty quantification.
  • This work builds upon previous research in property estimation and model projection.

Purpose of the Study:

  • To develop a novel method for estimating unknown numerical properties of physical objects (E) based on a limited set of measured properties (D).
  • To provide a framework for handling situations where the object's complete description requires significantly more parameters than available measurements.
  • To refine estimation techniques by incorporating Bayesian subjective probability for joint predictions with data errors.

Main Methods:

  • The study proposes a method for estimating P properties by assuming an upper bound (M) on the Hilbert norm of an orthogonal projection.
  • It simplifies notation from prior works (I and II) for clarity and broader applicability.
  • Bayesian subjective probability is explicitly applied to handle data errors and derive joint probability distributions for multiple predictions.

Main Results:

  • The paper presents a new approach to property estimation for complex physical systems (E) with limited data (D).
  • The method effectively estimates additional properties (P) by utilizing bounds on projected Hilbert norms.
  • The application of Bayesian probability enhances the handling of noisy data and joint predictions.

Conclusions:

  • The developed method offers a robust way to estimate multiple properties of complex physical objects even with incomplete data.
  • This approach improves predictive modeling by incorporating Hilbert norm bounds and Bayesian inference.
  • The work provides a simplified and explicit framework for advanced statistical estimation in physical sciences.

Related Experiment Videos