Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308505717> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4308505717 abstract "Camera images are ubiquitous in machine learning research. They also play a central role in the delivery of important services spanning medicine and environmental surveying. However, the application of machine learning models in these domains has been limited because of robustness concerns. A primary failure mode are performance drops due to differences between the training and deployment data. While there are methods to prospectively validate the robustness of machine learning models to such dataset drifts, existing approaches do not account for explicit models of the primary object of interest: the data. This limits our ability to study and understand the relationship between data generation and downstream machine learning model performance in a physically accurate manner. In this study, we demonstrate how to overcome this limitation by pairing traditional machine learning with physical optics to obtain explicit and differentiable data models. We demonstrate how such data models can be constructed for image data and used to control downstream machine learning model performance related to dataset drift. The findings are distilled into three applications. First, drift synthesis enables the controlled generation of physically faithful drift test cases to power model selection and targeted generalization. Second, the gradient connection between machine learning task model and data model allows advanced, precise tolerancing of task model sensitivity to changes in the data generation. These drift forensics can be used to precisely specify the acceptable data environments in which a task model may be run. Third, drift optimization opens up the possibility to create drifts that can help the task model learn better faster, effectively optimizing the data generating process itself. A guide to access the open code and datasets is available at https://github.com/aiaudit-org/raw2logit." @default.
- W4308505717 created "2022-11-12" @default.
- W4308505717 creator A5026451495 @default.
- W4308505717 creator A5028644233 @default.
- W4308505717 creator A5036144852 @default.
- W4308505717 creator A5045845258 @default.
- W4308505717 creator A5047327319 @default.
- W4308505717 creator A5053507592 @default.
- W4308505717 creator A5058200895 @default.
- W4308505717 creator A5061761068 @default.
- W4308505717 creator A5066411135 @default.
- W4308505717 creator A5070311555 @default.
- W4308505717 creator A5070918132 @default.
- W4308505717 creator A5073327508 @default.
- W4308505717 creator A5084976562 @default.
- W4308505717 date "2022-11-04" @default.
- W4308505717 modified "2023-10-10" @default.
- W4308505717 title "Data Models for Dataset Drift Controls in Machine Learning With Optical Images" @default.
- W4308505717 doi "https://doi.org/10.48550/arxiv.2211.02578" @default.
- W4308505717 hasPublicationYear "2022" @default.
- W4308505717 type Work @default.
- W4308505717 citedByCount "0" @default.
- W4308505717 crossrefType "posted-content" @default.
- W4308505717 hasAuthorship W4308505717A5026451495 @default.
- W4308505717 hasAuthorship W4308505717A5028644233 @default.
- W4308505717 hasAuthorship W4308505717A5036144852 @default.
- W4308505717 hasAuthorship W4308505717A5045845258 @default.
- W4308505717 hasAuthorship W4308505717A5047327319 @default.
- W4308505717 hasAuthorship W4308505717A5053507592 @default.
- W4308505717 hasAuthorship W4308505717A5058200895 @default.
- W4308505717 hasAuthorship W4308505717A5061761068 @default.
- W4308505717 hasAuthorship W4308505717A5066411135 @default.
- W4308505717 hasAuthorship W4308505717A5070311555 @default.
- W4308505717 hasAuthorship W4308505717A5070918132 @default.
- W4308505717 hasAuthorship W4308505717A5073327508 @default.
- W4308505717 hasAuthorship W4308505717A5084976562 @default.
- W4308505717 hasBestOaLocation W43085057171 @default.
- W4308505717 hasConcept C104317684 @default.
- W4308505717 hasConcept C119857082 @default.
- W4308505717 hasConcept C127413603 @default.
- W4308505717 hasConcept C154945302 @default.
- W4308505717 hasConcept C185592680 @default.
- W4308505717 hasConcept C201995342 @default.
- W4308505717 hasConcept C2780451532 @default.
- W4308505717 hasConcept C28006648 @default.
- W4308505717 hasConcept C41008148 @default.
- W4308505717 hasConcept C55493867 @default.
- W4308505717 hasConcept C60777511 @default.
- W4308505717 hasConcept C63479239 @default.
- W4308505717 hasConcept C89198739 @default.
- W4308505717 hasConcept C93959086 @default.
- W4308505717 hasConceptScore W4308505717C104317684 @default.
- W4308505717 hasConceptScore W4308505717C119857082 @default.
- W4308505717 hasConceptScore W4308505717C127413603 @default.
- W4308505717 hasConceptScore W4308505717C154945302 @default.
- W4308505717 hasConceptScore W4308505717C185592680 @default.
- W4308505717 hasConceptScore W4308505717C201995342 @default.
- W4308505717 hasConceptScore W4308505717C2780451532 @default.
- W4308505717 hasConceptScore W4308505717C28006648 @default.
- W4308505717 hasConceptScore W4308505717C41008148 @default.
- W4308505717 hasConceptScore W4308505717C55493867 @default.
- W4308505717 hasConceptScore W4308505717C60777511 @default.
- W4308505717 hasConceptScore W4308505717C63479239 @default.
- W4308505717 hasConceptScore W4308505717C89198739 @default.
- W4308505717 hasConceptScore W4308505717C93959086 @default.
- W4308505717 hasLocation W43085057171 @default.
- W4308505717 hasOpenAccess W4308505717 @default.
- W4308505717 hasPrimaryLocation W43085057171 @default.
- W4308505717 hasRelatedWork W2586223019 @default.
- W4308505717 hasRelatedWork W2604457090 @default.
- W4308505717 hasRelatedWork W2961085424 @default.
- W4308505717 hasRelatedWork W2964491809 @default.
- W4308505717 hasRelatedWork W3007679450 @default.
- W4308505717 hasRelatedWork W3199615610 @default.
- W4308505717 hasRelatedWork W3214672833 @default.
- W4308505717 hasRelatedWork W4294470244 @default.
- W4308505717 hasRelatedWork W4308505717 @default.
- W4308505717 hasRelatedWork W4319309271 @default.
- W4308505717 isParatext "false" @default.
- W4308505717 isRetracted "false" @default.
- W4308505717 workType "article" @default.