A robust adaptive Lasso estimator for the independent contamination model

Machkour, Jasin; Muma, Michael; Alt, Bastian; Zoubir, Abdelhak M.

doi:10.1016/j.sigpro.2020.107608

Cited by 6 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. , K, as inputs to a forward variable selection method such as the LARS algorithm [38], Lasso [39], elastic net [40], or related methods (e.g., [8], [9], [41]). These forward selection methods include no more than one variable in each iteration.…”

Section: B the T-rex Selectormentioning

confidence: 99%

False Discovery Rate Control for Fast Screening of Large-Scale Genomics Biobanks

Machkour,

Muma,

Palomar

2023

2023 IEEE Statistical Signal Processing Workshop (SSP)

View full text Add to dashboard Cite

Algorithms that ensure reproducible findings from large-scale, high-dimensional data are pivotal in numerous signal processing applications. In recent years, multivariate false discovery rate (FDR) controlling methods have emerged, providing guarantees even in high-dimensional settings where the number of variables surpasses the number of samples. However, these methods often fail to reliably control the FDR in the presence of highly dependent variable groups, a common characteristic in fields such as genomics and finance. To tackle this critical issue, we introduce a novel framework that accounts for general dependency structures. Our proposed dependency-aware T-Rex selector integrates hierarchical graphical models within the T-Rex framework to effectively harness the dependency structure among variables. Leveraging martingale theory, we prove that our variable penalization mechanism ensures FDR control. We further generalize the FDR-controlling framework by stating and proving a clear condition necessary for designing both graphical and non-graphical models that capture dependencies. Additionally, we formulate a fully integrated optimal calibration algorithm that concurrently determines the parameters of the graphical model and the T-Rex framework, such that the FDR is controlled while maximizing the number of selected variables. Numerical experiments and a breast cancer survival analysis usecase demonstrate that the proposed method is the only one among the state-of-the-art benchmark methods that controls the FDR and reliably detects genes that have been previously identified to be related to breast cancer. An open-source implementation is available within the R package TRexSelector on CRAN.

show abstract

Section: B the T-rex Selectormentioning

confidence: 99%

False Discovery Rate Control for Fast Screening of Large-Scale Genomics Biobanks

Machkour,

Muma,

Palomar

2023

2023 IEEE Statistical Signal Processing Workshop (SSP)

View full text Add to dashboard Cite

show abstract

“…One of the most famous such experiments is the European Tracer Experiment-ETEX [13,14]. A considerable amount of work evaluated the source inversion methods against data of ETEX [11,[15][16][17][18]. However, most of them concentrated on estimating the emission inventories and time parameters of the release, while only Tomas et al [11] reported on the estimated source location obtained after application of their model to the ETEX experiment.…”

Section: Introductionmentioning

confidence: 99%

Method of Source Identification Following an Accidental Release at an Unknown Location Using a Lagrangian Atmospheric Dispersion Model

Andronopoulos

Kovalets

2021

Atmosphere

View full text Add to dashboard Cite

A computationally efficient source inversion algorithm was developed and applied with the Lagrangian atmospheric dispersion model DIPCOT. In the process of source location estimation by minimizing a correlation-based cost function, the algorithm uses only the values of the time-integrated concentrations at the monitoring stations instead of all of the individual measurements in the full concentration-time series, resulting in a significant reduction in the number of integrations of the backward transport equations. Following the source location estimation the release start time, duration and emission rate are assessed. The developed algorithm was verified for the conditions of the ETEX-I (European Tracer Experiment—1st release). Using time-integrated measurements from all available stations, the distance between the estimated and true source location was 108 km. The estimated start time of the release was only about 1 h different from the true value, within the possible accuracy of estimate of this parameter. The estimated release duration was 21 h (the true value was 12 h). The estimated release rate was 4.28 g/s (the true value was 7.95 g/s). The estimated released mass almost perfectly fitted the true released mass (323.6 vs. 343.4 kg). It thus could be concluded that the developed algorithm is suitable for further integration in real-time decision support systems.

show abstract

Robust Machine Learning for Predicting Thermal Stability of Metal-Organic Framework

Azies,

Akrom,

Rustad

et al. 2024

Chemistry Africa

View full text Add to dashboard Cite

A robust adaptive Lasso estimator for the independent contamination model

Cited by 6 publications

References 25 publications

False Discovery Rate Control for Fast Screening of Large-Scale Genomics Biobanks

False Discovery Rate Control for Fast Screening of Large-Scale Genomics Biobanks

Method of Source Identification Following an Accidental Release at an Unknown Location Using a Lagrangian Atmospheric Dispersion Model

Robust Machine Learning for Predicting Thermal Stability of Metal-Organic Framework

Contact Info

Product

Resources

About