Version 1
: Received: 14 February 2022 / Approved: 21 February 2022 / Online: 21 February 2022 (14:13:26 CET)
Version 2
: Received: 7 June 2023 / Approved: 7 June 2023 / Online: 7 June 2023 (13:20:18 CEST)
How to cite:
Gogoshin, G.; Rodin, A. Minimum Uncertainty as Bayesian Network Model Selection Principle. Preprints2022, 2022020254. https://doi.org/10.20944/preprints202202.0254.v2
Gogoshin, G.; Rodin, A. Minimum Uncertainty as Bayesian Network Model Selection Principle. Preprints 2022, 2022020254. https://doi.org/10.20944/preprints202202.0254.v2
Gogoshin, G.; Rodin, A. Minimum Uncertainty as Bayesian Network Model Selection Principle. Preprints2022, 2022020254. https://doi.org/10.20944/preprints202202.0254.v2
APA Style
Gogoshin, G., & Rodin, A. (2023). Minimum Uncertainty as Bayesian Network Model Selection Principle. Preprints. https://doi.org/10.20944/preprints202202.0254.v2
Chicago/Turabian Style
Gogoshin, G. and Andrei Rodin. 2023 "Minimum Uncertainty as Bayesian Network Model Selection Principle" Preprints. https://doi.org/10.20944/preprints202202.0254.v2
Abstract
In thispaperstudy,we develop a Bayesian Network model selection principle thataddressaddressestheincommensurability of network features obtained from incongruous datasets and overcomes performanceirregularities of the Minimum Description Length model selection principle.This is achieved (i) byapproaching model evaluation as a classification problem, (ii) by estimating the effect that samplingerror has on the satisfiability of conditional independence criterion, as reflected by Mutual Information,and (iii) by utilizing this error estimate to penalize uncertainty in the Minimum Uncertainty (MU) modelselection principle. We validate our findings numerically and demonstrate the performance advantagesof the MU criterion. Finally, we illustrate the advantages of the new model evaluation framework on atRNA structural biology example.
Computer Science and Mathematics, Applied Mathematics
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received:
7 June 2023
Commenter:
Grigoriy Gogoshin
Commenter's Conflict of Interests:
Author
Comment:
The reasoning presented in the original manuscript was generalized and summarized into a model selection principle. The narrative has been substantially reworked and expanded. Consistency and coherence of the derivation of the bound on the statistical uncertainty were improved. Figures and tables were updated to reflect the changes. A structural biology application example was added --- dissection of the intra-tRNA-molecule residue (position) relationships.
Commenter: Grigoriy Gogoshin
Commenter's Conflict of Interests: Author