Stability selection
By Nicolai Meinshausen, P. Buehlmann. 2008
2,690 citations recorded by the source corpus.
Summary. Estimation of structure, such as in variable selection, graphical modelling or cluster analysis, is notoriously difficult, especially for high dimensional data. We introduce stability selection. It is based on subsampling in combination with (high dimensional) selection algorithms. As such, the method is extremely general and has a very wide range of applicability. Stability selection provides finite sample control for some error rates of false discoveries and hence a transparent principle to choose a proper amount of regularization for structure estimation. Variable selection and structure estimation improve markedly for a range of selection methods if stability selection is applied. We prove for the randomized lasso that stability selection will be variable selection consistent even if the necessary conditions for consistency of the original lasso method are violated. We demonstrate stability selection for variable selection and Gaussian graphical modelling, using real and simulated data.
Subfields: Artificial Intelligence, Statistics and Probability, Evaluation & Benchmarks, Statistical Learning Theory
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Peter Buhlmann
- Yoshua Bengio
- Michael I. Jordan
- Mohit Bansal
- Percy Liang
- Martin J. Wainwright
- Eric P. Xing
- Ruslan Salakhutdinov
In Statistics and Probability
- Peter Buhlmann
- Nicolai Meinshausen
- Robert Tibshirani
- Martin J. Wainwright
- Sara A. van de Geer
- Cun-Hui Zhang
- E.J. Candes
- Bin Yu
In Evaluation & Benchmarks
- Lucia Specia
- Kashif Shah
- Marco Turchi
- Matteo Negri
- Luc Van Gool
- Lesia Semenova
- Peter Buhlmann
- Jan Peters
In Statistical Learning Theory
- Sara A. van de Geer
- Peter Buhlmann
- Martin J. Wainwright
- Nicolai Meinshausen
- Robert Tibshirani
- Bin Yu
- Francis R. Bach
- Cun-Hui Zhang
Citations in the corpus
- Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “Algorithms like univariate selection, variance threshold, feature importance, correlation matrices (Saeys et al., 2007) or stability selection (Meinshausen & Bühlmann, 2010) are already integrated in modern AutoML frameworks (Thornton et al., 2013; Komer et al., 2014; Feurer et al., 2015a; Olson &…”
- Forward stability and model path selection (2021, neutral): “In the original SS formulation, Meinshausen and Buhlmann (2010) suggested that models built via forward selection be constructed on each resample to achieve a notion of stable variable selection.”
- All Models are Wrong, but Many are Useful: Learning a Variable’s Importance by Studying an Entire Class of Prediction Models Simultaneously (2018, background): “Meinshausen and Bühlmann (2010) estimate structural aspects of an underlying model (such as the variables included in that model) based on how stable those aspects are across a set of well-fitting models.”
- CAM: Causal Additive Models, high-dimensional order search and penalized regression (2013, methodology): “We also apply stability selection (Meinshausen and B¨uhlmann, 2010) to this data set.”
- On asymptotically optimal confidence regions and tests for high-dimensional models (2013, methodology): “of p-values and confidence intervals, and the procedure has been improved by using multiple sample splitting and aggregation of dependent p-values from multiple sample splits [32]. Stability selection [31] and its modification [41] provides another route to estimate error measures for false positive s”
- Order-independent constraint-based causal structure learning (2012, methodology): “Another possible solution for the order-dependence orientation issues would be to use stability selection (Meinshausen and Bühlmann, 2010) to find the most stable orientations among the runs.”
- p-Values for High-Dimensional Regression (2008, methodology): “Bach (2008) and Meinshausen and Bühlmann (2008) showed that for the Lasso, this leads to a consistent model selection procedure under fewer restrictions than for the nonbootstrap case.”
- Variable selection using pseudo-variables (2018, methodology): “, 2015), resampling methods (Hall et al., 2009; Meinshausen and Bühlmann, 2010; Feng and Yu, 2013; Sun et al., 2013; Shah and Samworth, 2013; Sabourin et al., 2015), and variable addition methods (Wu et al.”
- A Benchmark Dataset and Evaluation Methodology for Video Object Segmentation (2016, methodology): “To ensure robustness in the estimated graph we employ stability selection [30].”
- Finding and Leveraging Structure in Learning Problems (2012, methodology): “Recently, there have been some promising methods for tuning parameter selection based on stability and bootstrapping [Meinshausen and Bühlmann, 2010, Liu et al., 2010] which can select sparsistent models from finite samples.”
- The functional linear array model (2015, methodology): “Stability selection (Meinshausen and Bühlmann, 2010) can be used to improve variable selection.”
- SuSPect: Enhanced Prediction of Single Amino Acid Variant (SAV) Phenotype Using Network Features (2014, methodology): “In stability selection, feature selection is carried out on multiple subsets of 50% of the SAVs and the frequency at which each feature is chosen is used to determine important features [22].”
- Using Lasso for Predictor Selection and to Assuage Overfitting: A Method Long Overlooked in Behavioral Sciences (2015, methodology): “The problem is no different in the context of Lasso although Wasserman and Roeder (2009) and Meinshausen and Bühlmann (2010) have developed methods for computing Lasso p values by splitting the dataset.”
- Conditional transformation models (2012, methodology): “The stability selection procedure (Meinshausen and Bühlmann 2010) selected three variables that have an influence on the conditional distribution of the median housing value (MEDV), namely per capita crime (CRIM), average numbers of rooms per dwelling (RM), and percentage values of lower status popu”