Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386397730> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W4386397730 endingPage "119648" @default.
- W4386397730 startingPage "119648" @default.
- W4386397730 abstract "Broad learning system (BLS) has attracted widespread attention owing to its concise structure and efficient incremental learning based on ridge regression approximation of pseudo-inverse. However, BLS struggles with expensive computational costs and high memory usage when processing complex problems and training large networks due to the inverse operation of a large matrix. There has been some attempts to address this issue. In this work, we attempt to ameliorate the computational efficiency motivated by the state-of-the-art randomized iterative least squares solver. First, we improve the iteration manner of randomized extended Kaczmarz (REK) and then propose the iterated projection learning for training BLS without pseudo-inverse. The iterated projection learning is suitable for both offline and online learning scenarios and can adjust the model structure by incremental learning. Thanks to its linear convergence of least norm in expectation, the obtained model is expected to have better generalization ability without additional regularization procedure. Our model is found to maintain stable performance during incremental learning. Moreover, instead of grid search, we present an evolutionary bilevel programming (EBP) method with dispersion operator to further optimize the hyperparameters of the network structure. Numerical experiments indicate that the proposed methods can improve the efficiency, robustness, and generalization ability compared with pseudo-inverse and grid search based methods." @default.
- W4386397730 created "2023-09-03" @default.
- W4386397730 creator A5004496066 @default.
- W4386397730 creator A5017971063 @default.
- W4386397730 creator A5047357046 @default.
- W4386397730 creator A5083841661 @default.
- W4386397730 creator A5039773495 @default.
- W4386397730 date "2023-11-01" @default.
- W4386397730 modified "2023-10-16" @default.
- W4386397730 title "Pseudo Inverse versus Iterated Projection: Novel Learning Approach and its Application on Broad Learning System" @default.
- W4386397730 cites W1986278072 @default.
- W4386397730 cites W1990919278 @default.
- W4386397730 cites W1996640396 @default.
- W4386397730 cites W2012638612 @default.
- W4386397730 cites W2048305372 @default.
- W4386397730 cites W2056515450 @default.
- W4386397730 cites W2079325629 @default.
- W4386397730 cites W2135294617 @default.
- W4386397730 cites W2156601771 @default.
- W4386397730 cites W2738226240 @default.
- W4386397730 cites W2887576641 @default.
- W4386397730 cites W2889731581 @default.
- W4386397730 cites W2890126432 @default.
- W4386397730 cites W2890706287 @default.
- W4386397730 cites W2963227338 @default.
- W4386397730 cites W2980627566 @default.
- W4386397730 cites W2999246302 @default.
- W4386397730 cites W3044055626 @default.
- W4386397730 cites W3102157992 @default.
- W4386397730 cites W3115148317 @default.
- W4386397730 cites W3126008680 @default.
- W4386397730 cites W3171477298 @default.
- W4386397730 cites W3199466527 @default.
- W4386397730 cites W3207446465 @default.
- W4386397730 cites W4221067633 @default.
- W4386397730 cites W4225158741 @default.
- W4386397730 cites W4283524958 @default.
- W4386397730 cites W4287511352 @default.
- W4386397730 cites W4312547100 @default.
- W4386397730 cites W4367146573 @default.
- W4386397730 cites W4382247720 @default.
- W4386397730 doi "https://doi.org/10.1016/j.ins.2023.119648" @default.
- W4386397730 hasPublicationYear "2023" @default.
- W4386397730 type Work @default.
- W4386397730 citedByCount "0" @default.
- W4386397730 crossrefType "journal-article" @default.
- W4386397730 hasAuthorship W4386397730A5004496066 @default.
- W4386397730 hasAuthorship W4386397730A5017971063 @default.
- W4386397730 hasAuthorship W4386397730A5039773495 @default.
- W4386397730 hasAuthorship W4386397730A5047357046 @default.
- W4386397730 hasAuthorship W4386397730A5083841661 @default.
- W4386397730 hasConcept C10485038 @default.
- W4386397730 hasConcept C11413529 @default.
- W4386397730 hasConcept C119857082 @default.
- W4386397730 hasConcept C12267149 @default.
- W4386397730 hasConcept C126255220 @default.
- W4386397730 hasConcept C134306372 @default.
- W4386397730 hasConcept C140479938 @default.
- W4386397730 hasConcept C154945302 @default.
- W4386397730 hasConcept C187691185 @default.
- W4386397730 hasConcept C207467116 @default.
- W4386397730 hasConcept C2524010 @default.
- W4386397730 hasConcept C2778770139 @default.
- W4386397730 hasConcept C33923547 @default.
- W4386397730 hasConcept C41008148 @default.
- W4386397730 hasConcept C8642999 @default.
- W4386397730 hasConceptScore W4386397730C10485038 @default.
- W4386397730 hasConceptScore W4386397730C11413529 @default.
- W4386397730 hasConceptScore W4386397730C119857082 @default.
- W4386397730 hasConceptScore W4386397730C12267149 @default.
- W4386397730 hasConceptScore W4386397730C126255220 @default.
- W4386397730 hasConceptScore W4386397730C134306372 @default.
- W4386397730 hasConceptScore W4386397730C140479938 @default.
- W4386397730 hasConceptScore W4386397730C154945302 @default.
- W4386397730 hasConceptScore W4386397730C187691185 @default.
- W4386397730 hasConceptScore W4386397730C207467116 @default.
- W4386397730 hasConceptScore W4386397730C2524010 @default.
- W4386397730 hasConceptScore W4386397730C2778770139 @default.
- W4386397730 hasConceptScore W4386397730C33923547 @default.
- W4386397730 hasConceptScore W4386397730C41008148 @default.
- W4386397730 hasConceptScore W4386397730C8642999 @default.
- W4386397730 hasLocation W43863977301 @default.
- W4386397730 hasOpenAccess W4386397730 @default.
- W4386397730 hasPrimaryLocation W43863977301 @default.
- W4386397730 hasRelatedWork W3014750173 @default.
- W4386397730 hasRelatedWork W3139241485 @default.
- W4386397730 hasRelatedWork W3199608561 @default.
- W4386397730 hasRelatedWork W4281646320 @default.
- W4386397730 hasRelatedWork W4283697347 @default.
- W4386397730 hasRelatedWork W4287818966 @default.
- W4386397730 hasRelatedWork W4295309597 @default.
- W4386397730 hasRelatedWork W4298144215 @default.
- W4386397730 hasRelatedWork W4323894855 @default.
- W4386397730 hasRelatedWork W4381737452 @default.
- W4386397730 hasVolume "649" @default.
- W4386397730 isParatext "false" @default.
- W4386397730 isRetracted "false" @default.
- W4386397730 workType "article" @default.