Matches in SemOpenAlex for { <https://semopenalex.org/work/W2947368158> ?p ?o ?g. }
- W2947368158 abstract "Gaussian Processes (GPs) are powerful kernelized methods for non-parameteric regression used in many applications. However, their use is limited to a few thousand of training samples due to their cubic time complexity. In order to scale GPs to larger datasets, several sparse approximations based on so-called inducing points have been proposed in the literature. In this work we investigate the connection between a general class of sparse inducing point GP regression methods and Bayesian recursive estimation which enables Kalman Filter like updating for online learning. The majority of previous work has focused on the batch setting, in particular for learning the model parameters and the position of the inducing points, here instead we focus on training with mini-batches. By exploiting the Kalman filter formulation, we propose a novel approach that estimates such parameters by recursively propagating the analytical gradients of the posterior over mini-batches of the data. Compared to state of the art methods, our method keeps analytic updates for the mean and covariance of the posterior, thus reducing drastically the size of the optimization problem. We show that our method achieves faster convergence and superior performance compared to state of the art sequential Gaussian Process regression on synthetic GP as well as real-world data with up to a million of data samples." @default.
- W2947368158 created "2019-06-07" @default.
- W2947368158 creator A5007403791 @default.
- W2947368158 creator A5044312043 @default.
- W2947368158 creator A5062841940 @default.
- W2947368158 creator A5071722406 @default.
- W2947368158 date "2019-05-28" @default.
- W2947368158 modified "2023-09-27" @default.
- W2947368158 title "Recursive Estimation for Sparse Gaussian Process Regression" @default.
- W2947368158 cites W137285897 @default.
- W2947368158 cites W1506806321 @default.
- W2947368158 cites W1510444525 @default.
- W2947368158 cites W1512149552 @default.
- W2947368158 cites W151435321 @default.
- W2947368158 cites W1522301498 @default.
- W2947368158 cites W1571870753 @default.
- W2947368158 cites W1746819321 @default.
- W2947368158 cites W1777124189 @default.
- W2947368158 cites W1844625739 @default.
- W2947368158 cites W1886178467 @default.
- W2947368158 cites W1972537245 @default.
- W2947368158 cites W1999479532 @default.
- W2947368158 cites W2019130306 @default.
- W2947368158 cites W2021065610 @default.
- W2947368158 cites W2092766760 @default.
- W2947368158 cites W2099768828 @default.
- W2947368158 cites W2111611307 @default.
- W2947368158 cites W2123687908 @default.
- W2947368158 cites W2126877537 @default.
- W2947368158 cites W2129564505 @default.
- W2947368158 cites W2143880780 @default.
- W2947368158 cites W2161083632 @default.
- W2947368158 cites W2166851633 @default.
- W2947368158 cites W2222911609 @default.
- W2947368158 cites W2267434354 @default.
- W2947368158 cites W2396665160 @default.
- W2947368158 cites W2540189295 @default.
- W2947368158 cites W2568113389 @default.
- W2947368158 cites W2610838890 @default.
- W2947368158 cites W2617168490 @default.
- W2947368158 cites W2785523195 @default.
- W2947368158 cites W2809717747 @default.
- W2947368158 cites W2886463271 @default.
- W2947368158 cites W2952524552 @default.
- W2947368158 cites W2963433607 @default.
- W2947368158 cites W2963492326 @default.
- W2947368158 cites W2963662217 @default.
- W2947368158 cites W2963698657 @default.
- W2947368158 cites W2963779771 @default.
- W2947368158 hasPublicationYear "2019" @default.
- W2947368158 type Work @default.
- W2947368158 sameAs 2947368158 @default.
- W2947368158 citedByCount "1" @default.
- W2947368158 countsByYear W29473681582020 @default.
- W2947368158 crossrefType "posted-content" @default.
- W2947368158 hasAuthorship W2947368158A5007403791 @default.
- W2947368158 hasAuthorship W2947368158A5044312043 @default.
- W2947368158 hasAuthorship W2947368158A5062841940 @default.
- W2947368158 hasAuthorship W2947368158A5071722406 @default.
- W2947368158 hasConcept C105795698 @default.
- W2947368158 hasConcept C107673813 @default.
- W2947368158 hasConcept C11413529 @default.
- W2947368158 hasConcept C119857082 @default.
- W2947368158 hasConcept C121332964 @default.
- W2947368158 hasConcept C154945302 @default.
- W2947368158 hasConcept C157286648 @default.
- W2947368158 hasConcept C162324750 @default.
- W2947368158 hasConcept C163716315 @default.
- W2947368158 hasConcept C178650346 @default.
- W2947368158 hasConcept C2777303404 @default.
- W2947368158 hasConcept C33923547 @default.
- W2947368158 hasConcept C41008148 @default.
- W2947368158 hasConcept C50522688 @default.
- W2947368158 hasConcept C60229501 @default.
- W2947368158 hasConcept C61326573 @default.
- W2947368158 hasConcept C62520636 @default.
- W2947368158 hasConcept C76155785 @default.
- W2947368158 hasConcept C81692654 @default.
- W2947368158 hasConcept C83546350 @default.
- W2947368158 hasConceptScore W2947368158C105795698 @default.
- W2947368158 hasConceptScore W2947368158C107673813 @default.
- W2947368158 hasConceptScore W2947368158C11413529 @default.
- W2947368158 hasConceptScore W2947368158C119857082 @default.
- W2947368158 hasConceptScore W2947368158C121332964 @default.
- W2947368158 hasConceptScore W2947368158C154945302 @default.
- W2947368158 hasConceptScore W2947368158C157286648 @default.
- W2947368158 hasConceptScore W2947368158C162324750 @default.
- W2947368158 hasConceptScore W2947368158C163716315 @default.
- W2947368158 hasConceptScore W2947368158C178650346 @default.
- W2947368158 hasConceptScore W2947368158C2777303404 @default.
- W2947368158 hasConceptScore W2947368158C33923547 @default.
- W2947368158 hasConceptScore W2947368158C41008148 @default.
- W2947368158 hasConceptScore W2947368158C50522688 @default.
- W2947368158 hasConceptScore W2947368158C60229501 @default.
- W2947368158 hasConceptScore W2947368158C61326573 @default.
- W2947368158 hasConceptScore W2947368158C62520636 @default.
- W2947368158 hasConceptScore W2947368158C76155785 @default.
- W2947368158 hasConceptScore W2947368158C81692654 @default.
- W2947368158 hasConceptScore W2947368158C83546350 @default.
- W2947368158 hasLocation W29473681581 @default.