Matches in SemOpenAlex for { <https://semopenalex.org/work/W4321648564> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4321648564 abstract "Gradient boosting machines (GBMs) based on decision trees consistently demonstrate state-of-the-art results on regression and classification tasks with tabular data, often outperforming deep neural networks. However, these models do not provide well-calibrated predictive uncertainties, which prevents their use for decision making in high-risk applications. The Bayesian treatment is known to improve predictive uncertainty calibration, but previously proposed Bayesian GBM methods are either computationally expensive, or resort to crude approximations. Variational inference is often used to implement Bayesian neural networks, but is difficult to apply to GBMs, because the decision trees used as weak learners are non-differentiable. In this paper, we propose to implement Bayesian GBMs using variational inference with soft decision trees, a fully differentiable alternative to standard decision trees introduced by Irsoy et al. Our experiments demonstrate that variational soft trees and variational soft GBMs provide useful uncertainty estimates, while retaining good predictive performance. The proposed models show higher test likelihoods when compared to the state-of-the-art Bayesian GBMs in 7/10 tabular regression datasets and improved out-of-distribution detection in 5/10 datasets." @default.
- W4321648564 created "2023-02-24" @default.
- W4321648564 creator A5002232506 @default.
- W4321648564 creator A5011815288 @default.
- W4321648564 creator A5027386823 @default.
- W4321648564 creator A5031032237 @default.
- W4321648564 creator A5087483149 @default.
- W4321648564 date "2023-02-21" @default.
- W4321648564 modified "2023-09-26" @default.
- W4321648564 title "Variational Boosted Soft Trees" @default.
- W4321648564 doi "https://doi.org/10.48550/arxiv.2302.10706" @default.
- W4321648564 hasPublicationYear "2023" @default.
- W4321648564 type Work @default.
- W4321648564 citedByCount "0" @default.
- W4321648564 crossrefType "posted-content" @default.
- W4321648564 hasAuthorship W4321648564A5002232506 @default.
- W4321648564 hasAuthorship W4321648564A5011815288 @default.
- W4321648564 hasAuthorship W4321648564A5027386823 @default.
- W4321648564 hasAuthorship W4321648564A5031032237 @default.
- W4321648564 hasAuthorship W4321648564A5087483149 @default.
- W4321648564 hasBestOaLocation W43216485641 @default.
- W4321648564 hasConcept C105795698 @default.
- W4321648564 hasConcept C107673813 @default.
- W4321648564 hasConcept C119857082 @default.
- W4321648564 hasConcept C124101348 @default.
- W4321648564 hasConcept C134306372 @default.
- W4321648564 hasConcept C154945302 @default.
- W4321648564 hasConcept C160234255 @default.
- W4321648564 hasConcept C165838908 @default.
- W4321648564 hasConcept C169258074 @default.
- W4321648564 hasConcept C202615002 @default.
- W4321648564 hasConcept C2776214188 @default.
- W4321648564 hasConcept C33923547 @default.
- W4321648564 hasConcept C41008148 @default.
- W4321648564 hasConcept C46686674 @default.
- W4321648564 hasConcept C50644808 @default.
- W4321648564 hasConcept C70153297 @default.
- W4321648564 hasConcept C83546350 @default.
- W4321648564 hasConcept C84525736 @default.
- W4321648564 hasConceptScore W4321648564C105795698 @default.
- W4321648564 hasConceptScore W4321648564C107673813 @default.
- W4321648564 hasConceptScore W4321648564C119857082 @default.
- W4321648564 hasConceptScore W4321648564C124101348 @default.
- W4321648564 hasConceptScore W4321648564C134306372 @default.
- W4321648564 hasConceptScore W4321648564C154945302 @default.
- W4321648564 hasConceptScore W4321648564C160234255 @default.
- W4321648564 hasConceptScore W4321648564C165838908 @default.
- W4321648564 hasConceptScore W4321648564C169258074 @default.
- W4321648564 hasConceptScore W4321648564C202615002 @default.
- W4321648564 hasConceptScore W4321648564C2776214188 @default.
- W4321648564 hasConceptScore W4321648564C33923547 @default.
- W4321648564 hasConceptScore W4321648564C41008148 @default.
- W4321648564 hasConceptScore W4321648564C46686674 @default.
- W4321648564 hasConceptScore W4321648564C50644808 @default.
- W4321648564 hasConceptScore W4321648564C70153297 @default.
- W4321648564 hasConceptScore W4321648564C83546350 @default.
- W4321648564 hasConceptScore W4321648564C84525736 @default.
- W4321648564 hasLocation W43216485641 @default.
- W4321648564 hasOpenAccess W4321648564 @default.
- W4321648564 hasPrimaryLocation W43216485641 @default.
- W4321648564 hasRelatedWork W3036633074 @default.
- W4321648564 hasRelatedWork W3159988495 @default.
- W4321648564 hasRelatedWork W3200719183 @default.
- W4321648564 hasRelatedWork W3212730154 @default.
- W4321648564 hasRelatedWork W4200057378 @default.
- W4321648564 hasRelatedWork W4206556944 @default.
- W4321648564 hasRelatedWork W4288057626 @default.
- W4321648564 hasRelatedWork W4293069612 @default.
- W4321648564 hasRelatedWork W4296081764 @default.
- W4321648564 hasRelatedWork W4320854072 @default.
- W4321648564 isParatext "false" @default.
- W4321648564 isRetracted "false" @default.
- W4321648564 workType "article" @default.