dorsal/arxiv
View SchemaAlgorithm for Model Validation: Theory and Applications
| Authors | D. Sornette, A. B. Davis, K. Ide, K. R. Vixie, V. Pisarenko, J. R. Kamm |
|---|---|
| Categories | |
| ArXiv ID | physics/0511219 |
| URL | https://arxiv.org/abs/physics/0511219 |
| DOI | 10.1073/pnas.0611677104 |
| Journal | Proc. Nat. Acad. Sci. USA 104 (16), 6562-6567 (2007) |
Abstract
Validation is often defined as the process of determining the degree to which a model is an accurate representation of the real world from the perspective of its intended uses. Validation is crucial as industries and governments depend increasingly on predictions by computer models to justify their decisions. We propose to formulate the validation of a given model as an iterative construction process that mimics the often implicit process occurring in the minds of scientists. We offer a formal representation of the progressive build-up of trust in the model. We thus replace static claims on the impossibility of validating a given model by a dynamic process of constructive approximation. This approach is better adapted to the fuzzy, coarse-grained nature of validation. Our procedure factors in the degree of redundancy versus novelty of the experiments used for validation as well as the degree to which the model predicts the observations. We illustrate the new methodology first with the maturation of Quantum Mechanics as the arguably best established physics theory and then with several concrete examples drawn from some of our primary scientific interests: a cellular automaton model for earthquakes, a multifractal random walk model for financial time series, an anomalous diffusion model for solar radiation transport in the cloudy atmosphere, and a computational fluid dynamics code for the Richtmyer-Meshkov instability.
{
"annotation_id": "a665f67c-f965-49c3-9aa9-17042739262b",
"date_created": "2026-03-02T18:01:03.793000Z",
"date_modified": "2026-03-02T18:01:03.793000Z",
"file_hash": "d23d08528e40f071b6c523ba5a7fab3fe467955cebb80963163c51a487e8a731",
"private": false,
"record": {
"abstract": "Validation is often defined as the process of determining the degree to which\na model is an accurate representation of the real world from the perspective of\nits intended uses. Validation is crucial as industries and governments depend\nincreasingly on predictions by computer models to justify their decisions. We\npropose to formulate the validation of a given model as an iterative\nconstruction process that mimics the often implicit process occurring in the\nminds of scientists. We offer a formal representation of the progressive\nbuild-up of trust in the model. We thus replace static claims on the\nimpossibility of validating a given model by a dynamic process of constructive\napproximation. This approach is better adapted to the fuzzy, coarse-grained\nnature of validation. Our procedure factors in the degree of redundancy versus\nnovelty of the experiments used for validation as well as the degree to which\nthe model predicts the observations. We illustrate the new methodology first\nwith the maturation of Quantum Mechanics as the arguably best established\nphysics theory and then with several concrete examples drawn from some of our\nprimary scientific interests: a cellular automaton model for earthquakes, a\nmultifractal random walk model for financial time series, an anomalous\ndiffusion model for solar radiation transport in the cloudy atmosphere, and a\ncomputational fluid dynamics code for the Richtmyer-Meshkov instability.",
"arxiv_id": "physics/0511219",
"authors": [
"D. Sornette",
"A. B. Davis",
"K. Ide",
"K. R. Vixie",
"V. Pisarenko",
"J. R. Kamm"
],
"categories": [
"physics.data-an",
"physics.comp-ph"
],
"doi": "10.1073/pnas.0611677104",
"journal_ref": "Proc. Nat. Acad. Sci. USA 104 (16), 6562-6567 (2007)",
"title": "Algorithm for Model Validation: Theory and Applications",
"url": "https://arxiv.org/abs/physics/0511219"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "4ac95698-c102-46ea-888a-761e28b27276",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}