Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210680003> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3210680003 endingPage "10" @default.
- W3210680003 startingPage "1" @default.
- W3210680003 abstract "The Gauss Markov theorem states that the weighted least squares estimator is a linear minimum variance unbiased estimation (MVUE) in linear models. In this paper, we take a first step towards extending this result to non-linear settings via deep learning with bias constraints. The classical approach to designing non-linear MVUEs is through maximum likelihood estimation (MLE) which often involves real-time computationally challenging optimizations. On the other hand, deep learning methods allow for non-linear estimators with fixed computational complexity. Learning based estimators perform optimally on average with respect to their training set but may suffer from significant bias in other parameters. To avoid this, we propose to add a simple bias constraint to the loss function, resulting in an estimator we refer to as Bias Constrained Estimator (BCE). We prove that this yields asymptotic MVUEs that behave similarly to the classical MLEs and asymptotically attain the Cramer Rao bound. We demonstrate the advantages of our approach in the context of signal to noise ratio estimation as well as covariance estimation. A second motivation to BCE is in applications where multiple estimates of the same unknown are averaged for improved performance. Examples include distributed sensor networks and data augmentation in test-time. In such applications, we show that BCE leads to asymptotically consistent estimators." @default.
- W3210680003 created "2021-11-08" @default.
- W3210680003 creator A5005913897 @default.
- W3210680003 creator A5055644922 @default.
- W3210680003 creator A5091732723 @default.
- W3210680003 date "2023-01-01" @default.
- W3210680003 modified "2023-10-16" @default.
- W3210680003 title "Learning to Estimate Without Bias" @default.
- W3210680003 cites W1554944419 @default.
- W3210680003 cites W1686810756 @default.
- W3210680003 cites W1821462560 @default.
- W3210680003 cites W1987368804 @default.
- W3210680003 cites W2097032753 @default.
- W3210680003 cites W2097117768 @default.
- W3210680003 cites W2108475251 @default.
- W3210680003 cites W2114863795 @default.
- W3210680003 cites W2134480188 @default.
- W3210680003 cites W2135341600 @default.
- W3210680003 cites W2145614561 @default.
- W3210680003 cites W2149310890 @default.
- W3210680003 cites W2157094190 @default.
- W3210680003 cites W2160747081 @default.
- W3210680003 cites W2245830722 @default.
- W3210680003 cites W2282391807 @default.
- W3210680003 cites W2294604761 @default.
- W3210680003 cites W2792251271 @default.
- W3210680003 cites W2806475032 @default.
- W3210680003 cites W2952399630 @default.
- W3210680003 cites W2953494151 @default.
- W3210680003 cites W2963323244 @default.
- W3210680003 cites W2970277617 @default.
- W3210680003 cites W2979839315 @default.
- W3210680003 cites W3021094251 @default.
- W3210680003 cites W3100944100 @default.
- W3210680003 cites W3111382075 @default.
- W3210680003 cites W3115024154 @default.
- W3210680003 cites W3132954326 @default.
- W3210680003 cites W3166374981 @default.
- W3210680003 cites W3170865782 @default.
- W3210680003 cites W3208643254 @default.
- W3210680003 cites W607505555 @default.
- W3210680003 doi "https://doi.org/10.1109/tsp.2023.3284372" @default.
- W3210680003 hasPublicationYear "2023" @default.
- W3210680003 type Work @default.
- W3210680003 sameAs 3210680003 @default.
- W3210680003 citedByCount "0" @default.
- W3210680003 crossrefType "journal-article" @default.
- W3210680003 hasAuthorship W3210680003A5005913897 @default.
- W3210680003 hasAuthorship W3210680003A5055644922 @default.
- W3210680003 hasAuthorship W3210680003A5091732723 @default.
- W3210680003 hasBestOaLocation W32106800032 @default.
- W3210680003 hasConcept C105795698 @default.
- W3210680003 hasConcept C11413529 @default.
- W3210680003 hasConcept C126255220 @default.
- W3210680003 hasConcept C134306372 @default.
- W3210680003 hasConcept C151730666 @default.
- W3210680003 hasConcept C178650346 @default.
- W3210680003 hasConcept C185429906 @default.
- W3210680003 hasConcept C2779343474 @default.
- W3210680003 hasConcept C28826006 @default.
- W3210680003 hasConcept C33923547 @default.
- W3210680003 hasConcept C41008148 @default.
- W3210680003 hasConcept C77553402 @default.
- W3210680003 hasConcept C86803240 @default.
- W3210680003 hasConceptScore W3210680003C105795698 @default.
- W3210680003 hasConceptScore W3210680003C11413529 @default.
- W3210680003 hasConceptScore W3210680003C126255220 @default.
- W3210680003 hasConceptScore W3210680003C134306372 @default.
- W3210680003 hasConceptScore W3210680003C151730666 @default.
- W3210680003 hasConceptScore W3210680003C178650346 @default.
- W3210680003 hasConceptScore W3210680003C185429906 @default.
- W3210680003 hasConceptScore W3210680003C2779343474 @default.
- W3210680003 hasConceptScore W3210680003C28826006 @default.
- W3210680003 hasConceptScore W3210680003C33923547 @default.
- W3210680003 hasConceptScore W3210680003C41008148 @default.
- W3210680003 hasConceptScore W3210680003C77553402 @default.
- W3210680003 hasConceptScore W3210680003C86803240 @default.
- W3210680003 hasLocation W32106800031 @default.
- W3210680003 hasLocation W32106800032 @default.
- W3210680003 hasOpenAccess W3210680003 @default.
- W3210680003 hasPrimaryLocation W32106800031 @default.
- W3210680003 hasRelatedWork W1969481295 @default.
- W3210680003 hasRelatedWork W1982639832 @default.
- W3210680003 hasRelatedWork W1982914560 @default.
- W3210680003 hasRelatedWork W2088931338 @default.
- W3210680003 hasRelatedWork W2118585214 @default.
- W3210680003 hasRelatedWork W2152044198 @default.
- W3210680003 hasRelatedWork W2152704622 @default.
- W3210680003 hasRelatedWork W2618513040 @default.
- W3210680003 hasRelatedWork W2743834469 @default.
- W3210680003 hasRelatedWork W3157804329 @default.
- W3210680003 isParatext "false" @default.
- W3210680003 isRetracted "false" @default.
- W3210680003 magId "3210680003" @default.
- W3210680003 workType "article" @default.