MétaCan
Menu
Back to cohort
Record W7028975089

High-dimensional graphical models for noisy data

2023· dissertation· en· W7028975089 on OpenAlexafffund

Bibliographic record

VenueeScholarship@McGill (McGill) · 2023
Typedissertation
Languageen
FieldComputer Science
TopicTopological and Geometric Data Analysis
Canadian institutionsMcGill University
FundersNatural Sciences and Engineering Research Council of CanadaMcGill University
KeywordsEstimatorCovarianceCovariance matrixEstimation of covariance matricesBounded functionConsistency (knowledge bases)Multiplicative functionLasso (programming language)Observational errorMissing data
DOInot available

Abstract

fetched live from OpenAlex

The problem of estimating the inverse covariance or precision matrix for graphical models under a high-dimensional setting is a well-known challenge in modern statistics.Numerous theoretical and applied works have been proposed to date, particularly when the data are fully observed and follow a multivariate normal distribution.However, in the presence of measurement errors, such as additive or multiplicative errors, different surrogate estimates have been suggested in the literature to obtain unbiased estimates of the true covariance matrix.Unfortunately, these surrogate estimators may not necessarily be positive semi-definite, leading to a non-convex objective function.To address this issue, the surrogate estimators can be projected onto the nearest positive semi-definite matrix, transforming the objective function into a convex problem.While consistency bounds for tail deviations of the estimated and true covariance matrix have been well-studied for fully observed data with sub-Gaussian distributions or bounded moments, such bounds have not been established for the presence of measurement errors.Therefore, the first part of this thesis focuses on developing consistency bounds for random variables that are sub-Gaussian or have bounded moments in the presence of additive or multiplicative measurement errors.We also perform simulation studies and real data analysis to compare the performance of the covariance projection method with existing methods for precision matrix estimation for corrupted data.i Next, we address the problem of joint estimation of regression coefficients and precision matrix in the presence of missing data, a common issue in genetics.We restrict our attention to the scenario where both the data and measurement error are sub-Gaussian.We employ similar techniques to project the surrogate estimate of the sample covariance matrix to ensure convexity of the objective function and derive consistency bounds.Additionally, we conduct simulation studies to compare our method with existing approaches.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.014
metaresearch head score (Gemma)0.056
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.014
Threshold uncertainty score0.072

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0140.056
Meta-epidemiology (narrow)0.0020.002
Meta-epidemiology (broad)0.0040.003
Bibliometrics0.0040.005
Science and technology studies0.0010.005
Scholarly communication0.0050.005
Open science0.0050.005
Research integrity0.0040.006
Insufficient payload (model declined to judge)0.0050.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.050
GPT teacher head0.278
Teacher spread0.227 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2023
Admission routes2
Has abstractyes

Explore more

Same venueeScholarship@McGill (McGill)Same topicTopological and Geometric Data AnalysisFrench-language works237,207