Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386561606> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4386561606 abstract "Estimating the transferability of publicly available pretrained models to a target task has assumed an important place for transfer learning tasks in recent years. Existing efforts propose metrics that allow a user to choose one model from a pool of pre-trained models without having to fine-tune each model individually and identify one explicitly. With the growth in the number of available pre-trained models and the popularity of model ensembles, it also becomes essential to study the transferability of multiple-source models for a given target task. The few existing efforts study transferability in such multi-source ensemble settings using just the outputs of the classification layer and neglect possible domain or task mismatch. Moreover, they overlook the most important factor while selecting the source models, viz., the cohesiveness factor between them, which can impact the performance and confidence in the prediction of the ensemble. To address these gaps, we propose a novel Optimal tranSport-based suBmOdular tRaNsferability metric (OSBORN) to estimate the transferability of an ensemble of models to a downstream task. OSBORN collectively accounts for image domain difference, task difference, and cohesiveness of models in the ensemble to provide reliable estimates of transferability. We gauge the performance of OSBORN on both image classification and semantic segmentation tasks. Our setup includes 28 source datasets, 11 target datasets, 5 model architectures, and 2 pre-training methods. We benchmark our method against current state-of-the-art metrics MS-LEEP and E-LEEP, and outperform them consistently using the proposed approach." @default.
- W4386561606 created "2023-09-09" @default.
- W4386561606 creator A5016441741 @default.
- W4386561606 creator A5034303942 @default.
- W4386561606 creator A5038020125 @default.
- W4386561606 creator A5043824731 @default.
- W4386561606 creator A5066394402 @default.
- W4386561606 creator A5092830332 @default.
- W4386561606 date "2023-09-05" @default.
- W4386561606 modified "2023-09-27" @default.
- W4386561606 title "Building a Winning Team: Selecting Source Model Ensembles using a Submodular Transferability Estimation Approach" @default.
- W4386561606 doi "https://doi.org/10.48550/arxiv.2309.02429" @default.
- W4386561606 hasPublicationYear "2023" @default.
- W4386561606 type Work @default.
- W4386561606 citedByCount "0" @default.
- W4386561606 crossrefType "posted-content" @default.
- W4386561606 hasAuthorship W4386561606A5016441741 @default.
- W4386561606 hasAuthorship W4386561606A5034303942 @default.
- W4386561606 hasAuthorship W4386561606A5038020125 @default.
- W4386561606 hasAuthorship W4386561606A5043824731 @default.
- W4386561606 hasAuthorship W4386561606A5066394402 @default.
- W4386561606 hasAuthorship W4386561606A5092830332 @default.
- W4386561606 hasBestOaLocation W43865616061 @default.
- W4386561606 hasConcept C114614502 @default.
- W4386561606 hasConcept C119857082 @default.
- W4386561606 hasConcept C119898033 @default.
- W4386561606 hasConcept C124101348 @default.
- W4386561606 hasConcept C13280743 @default.
- W4386561606 hasConcept C140331021 @default.
- W4386561606 hasConcept C154945302 @default.
- W4386561606 hasConcept C162324750 @default.
- W4386561606 hasConcept C176217482 @default.
- W4386561606 hasConcept C178621042 @default.
- W4386561606 hasConcept C185798385 @default.
- W4386561606 hasConcept C187736073 @default.
- W4386561606 hasConcept C205649164 @default.
- W4386561606 hasConcept C21547014 @default.
- W4386561606 hasConcept C2780451532 @default.
- W4386561606 hasConcept C33923547 @default.
- W4386561606 hasConcept C41008148 @default.
- W4386561606 hasConcept C61272859 @default.
- W4386561606 hasConceptScore W4386561606C114614502 @default.
- W4386561606 hasConceptScore W4386561606C119857082 @default.
- W4386561606 hasConceptScore W4386561606C119898033 @default.
- W4386561606 hasConceptScore W4386561606C124101348 @default.
- W4386561606 hasConceptScore W4386561606C13280743 @default.
- W4386561606 hasConceptScore W4386561606C140331021 @default.
- W4386561606 hasConceptScore W4386561606C154945302 @default.
- W4386561606 hasConceptScore W4386561606C162324750 @default.
- W4386561606 hasConceptScore W4386561606C176217482 @default.
- W4386561606 hasConceptScore W4386561606C178621042 @default.
- W4386561606 hasConceptScore W4386561606C185798385 @default.
- W4386561606 hasConceptScore W4386561606C187736073 @default.
- W4386561606 hasConceptScore W4386561606C205649164 @default.
- W4386561606 hasConceptScore W4386561606C21547014 @default.
- W4386561606 hasConceptScore W4386561606C2780451532 @default.
- W4386561606 hasConceptScore W4386561606C33923547 @default.
- W4386561606 hasConceptScore W4386561606C41008148 @default.
- W4386561606 hasConceptScore W4386561606C61272859 @default.
- W4386561606 hasLocation W43865616061 @default.
- W4386561606 hasOpenAccess W4386561606 @default.
- W4386561606 hasPrimaryLocation W43865616061 @default.
- W4386561606 hasRelatedWork W1485630101 @default.
- W4386561606 hasRelatedWork W2254744191 @default.
- W4386561606 hasRelatedWork W2498017833 @default.
- W4386561606 hasRelatedWork W2983785000 @default.
- W4386561606 hasRelatedWork W3136979370 @default.
- W4386561606 hasRelatedWork W4226365389 @default.
- W4386561606 hasRelatedWork W4281560664 @default.
- W4386561606 hasRelatedWork W4285741730 @default.
- W4386561606 hasRelatedWork W4383987268 @default.
- W4386561606 hasRelatedWork W3214927170 @default.
- W4386561606 isParatext "false" @default.
- W4386561606 isRetracted "false" @default.
- W4386561606 workType "article" @default.