Matches in SemOpenAlex for { <https://semopenalex.org/work/W4281647557> ?p ?o ?g. }
Showing items 1 to 57 of
57
with 100 items per page.
- W4281647557 abstract "To promote new scientific discoveries from complex data sets, feature importance inference has been a long-standing statistical problem. Instead of testing for parameters that are only interpretable for specific models, there has been increasing interest in model-agnostic methods, often in the form of feature occlusion or leave-one-covariate-out (LOCO) inference. Existing approaches often make distributional assumptions, which can be difficult to verify in practice, or require model refitting and data splitting, which are computationally intensive and lead to losses in power. In this work, we develop a novel, mostly model-agnostic and distribution-free inference framework for feature importance that is computationally efficient and statistically powerful. Our approach is fast as we avoid model refitting by leveraging a form of random observation and feature subsampling called minipatch ensembles; this approach also improves statistical power by avoiding data splitting. Our framework can be applied on tabular data and with any machine learning algorithm, together with minipatch ensembles, for regression and classification tasks. Despite the dependencies induced by using minipatch ensembles, we show that our approach provides asymptotic coverage for the feature importance score of any model under mild assumptions. Finally, our same procedure can also be leveraged to provide valid confidence intervals for predictions, hence providing fast, simultaneous quantification of the uncertainty of both predictions and feature importance. We validate our intervals on a series of synthetic and real data examples, including non-linear settings, showing that our approach detects the correct important features and exhibits many computational and statistical advantages over existing methods." @default.
- W4281647557 created "2022-06-13" @default.
- W4281647557 creator A5014971614 @default.
- W4281647557 creator A5018970288 @default.
- W4281647557 creator A5048412925 @default.
- W4281647557 date "2022-06-04" @default.
- W4281647557 modified "2023-09-29" @default.
- W4281647557 title "Model-Agnostic Confidence Intervals for Feature Importance: A Fast and Powerful Approach Using Minipatch Ensembles" @default.
- W4281647557 doi "https://doi.org/10.48550/arxiv.2206.02088" @default.
- W4281647557 hasPublicationYear "2022" @default.
- W4281647557 type Work @default.
- W4281647557 citedByCount "0" @default.
- W4281647557 crossrefType "posted-content" @default.
- W4281647557 hasAuthorship W4281647557A5014971614 @default.
- W4281647557 hasAuthorship W4281647557A5018970288 @default.
- W4281647557 hasAuthorship W4281647557A5048412925 @default.
- W4281647557 hasBestOaLocation W42816475571 @default.
- W4281647557 hasConcept C105795698 @default.
- W4281647557 hasConcept C119043178 @default.
- W4281647557 hasConcept C119857082 @default.
- W4281647557 hasConcept C124101348 @default.
- W4281647557 hasConcept C134261354 @default.
- W4281647557 hasConcept C138885662 @default.
- W4281647557 hasConcept C154945302 @default.
- W4281647557 hasConcept C2776214188 @default.
- W4281647557 hasConcept C2776401178 @default.
- W4281647557 hasConcept C33923547 @default.
- W4281647557 hasConcept C41008148 @default.
- W4281647557 hasConcept C41895202 @default.
- W4281647557 hasConceptScore W4281647557C105795698 @default.
- W4281647557 hasConceptScore W4281647557C119043178 @default.
- W4281647557 hasConceptScore W4281647557C119857082 @default.
- W4281647557 hasConceptScore W4281647557C124101348 @default.
- W4281647557 hasConceptScore W4281647557C134261354 @default.
- W4281647557 hasConceptScore W4281647557C138885662 @default.
- W4281647557 hasConceptScore W4281647557C154945302 @default.
- W4281647557 hasConceptScore W4281647557C2776214188 @default.
- W4281647557 hasConceptScore W4281647557C2776401178 @default.
- W4281647557 hasConceptScore W4281647557C33923547 @default.
- W4281647557 hasConceptScore W4281647557C41008148 @default.
- W4281647557 hasConceptScore W4281647557C41895202 @default.
- W4281647557 hasLocation W42816475571 @default.
- W4281647557 hasOpenAccess W4281647557 @default.
- W4281647557 hasPrimaryLocation W42816475571 @default.
- W4281647557 hasRelatedWork W2511279186 @default.
- W4281647557 hasRelatedWork W2981347089 @default.
- W4281647557 hasRelatedWork W3001657888 @default.
- W4281647557 hasRelatedWork W3098026267 @default.
- W4281647557 hasRelatedWork W3175389394 @default.
- W4281647557 hasRelatedWork W4221162807 @default.
- W4281647557 hasRelatedWork W4225834258 @default.
- W4281647557 hasRelatedWork W4226057380 @default.
- W4281647557 hasRelatedWork W4255939854 @default.
- W4281647557 hasRelatedWork W4317940208 @default.
- W4281647557 isParatext "false" @default.
- W4281647557 isRetracted "false" @default.
- W4281647557 workType "article" @default.