SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384829168> ?p ?o ?g. }

Showing items 1 to 100 of ±103 with 100 items per page.

W4384829168 endingPage "6511" @default.
W4384829168 startingPage "6511" @default.
W4384829168 abstract "The study of Domain Generalization (DG) has gained considerable momentum in the Machine Learning (ML) field. Human Activity Recognition (HAR) inherently encompasses diverse domains (e.g., users, devices, or datasets), rendering it an ideal testbed for exploring Domain Generalization. Building upon recent work, this paper investigates the application of regularization methods to bridge the generalization gap between traditional models based on handcrafted features and deep neural networks. We apply various regularizers, including sparse training, Mixup, Distributionally Robust Optimization (DRO), and Sharpness-Aware Minimization (SAM), to deep learning models and assess their performance in Out-of-Distribution (OOD) settings across multiple domains using homogenized public datasets. Our results show that Mixup and SAM are the best-performing regularizers. However, they are unable to match the performance of models based on handcrafted features. This suggests that while regularization techniques can improve OOD robustness to some extent, handcrafted features remain superior for domain generalization in HAR tasks." @default.
W4384829168 created "2023-07-21" @default.
W4384829168 creator A5061814910 @default.
W4384829168 creator A5080343600 @default.
W4384829168 creator A5080816268 @default.
W4384829168 creator A5084185357 @default.
W4384829168 creator A5092502166 @default.
W4384829168 date "2023-07-19" @default.
W4384829168 modified "2023-09-26" @default.
W4384829168 title "Exploring Regularization Methods for Domain Generalization in Accelerometer-Based Human Activity Recognition" @default.
W4384829168 cites W1852728451 @default.
W4384829168 cites W2096680959 @default.
W4384829168 cites W2100085204 @default.
W4384829168 cites W2129793335 @default.
W4384829168 cites W2194775991 @default.
W4384829168 cites W2247209766 @default.
W4384829168 cites W2340025709 @default.
W4384829168 cites W2577415454 @default.
W4384829168 cites W2736191430 @default.
W4384829168 cites W2883074505 @default.
W4384829168 cites W2914106530 @default.
W4384829168 cites W2929050222 @default.
W4384829168 cites W2963563124 @default.
W4384829168 cites W2988019621 @default.
W4384829168 cites W3012919764 @default.
W4384829168 cites W3083016735 @default.
W4384829168 cites W3135542633 @default.
W4384829168 cites W3136077883 @default.
W4384829168 cites W3210766530 @default.
W4384829168 cites W3214868319 @default.
W4384829168 cites W4225166090 @default.
W4384829168 cites W4283722463 @default.
W4384829168 cites W4289517877 @default.
W4384829168 cites W4289639938 @default.
W4384829168 cites W4297491167 @default.
W4384829168 cites W4307823382 @default.
W4384829168 doi "https://doi.org/10.3390/s23146511" @default.
W4384829168 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37514805" @default.
W4384829168 hasPublicationYear "2023" @default.
W4384829168 type Work @default.
W4384829168 citedByCount "0" @default.
W4384829168 crossrefType "journal-article" @default.
W4384829168 hasAuthorship W4384829168A5061814910 @default.
W4384829168 hasAuthorship W4384829168A5080343600 @default.
W4384829168 hasAuthorship W4384829168A5080816268 @default.
W4384829168 hasAuthorship W4384829168A5084185357 @default.
W4384829168 hasAuthorship W4384829168A5092502166 @default.
W4384829168 hasBestOaLocation W43848291681 @default.
W4384829168 hasConcept C104317684 @default.
W4384829168 hasConcept C108583219 @default.
W4384829168 hasConcept C119857082 @default.
W4384829168 hasConcept C134306372 @default.
W4384829168 hasConcept C154945302 @default.
W4384829168 hasConcept C177148314 @default.
W4384829168 hasConcept C185592680 @default.
W4384829168 hasConcept C205711294 @default.
W4384829168 hasConcept C2776135515 @default.
W4384829168 hasConcept C2984842247 @default.
W4384829168 hasConcept C31258907 @default.
W4384829168 hasConcept C31395832 @default.
W4384829168 hasConcept C33923547 @default.
W4384829168 hasConcept C41008148 @default.
W4384829168 hasConcept C55493867 @default.
W4384829168 hasConcept C63479239 @default.
W4384829168 hasConceptScore W4384829168C104317684 @default.
W4384829168 hasConceptScore W4384829168C108583219 @default.
W4384829168 hasConceptScore W4384829168C119857082 @default.
W4384829168 hasConceptScore W4384829168C134306372 @default.
W4384829168 hasConceptScore W4384829168C154945302 @default.
W4384829168 hasConceptScore W4384829168C177148314 @default.
W4384829168 hasConceptScore W4384829168C185592680 @default.
W4384829168 hasConceptScore W4384829168C205711294 @default.
W4384829168 hasConceptScore W4384829168C2776135515 @default.
W4384829168 hasConceptScore W4384829168C2984842247 @default.
W4384829168 hasConceptScore W4384829168C31258907 @default.
W4384829168 hasConceptScore W4384829168C31395832 @default.
W4384829168 hasConceptScore W4384829168C33923547 @default.
W4384829168 hasConceptScore W4384829168C41008148 @default.
W4384829168 hasConceptScore W4384829168C55493867 @default.
W4384829168 hasConceptScore W4384829168C63479239 @default.
W4384829168 hasFunder F4320334779 @default.
W4384829168 hasIssue "14" @default.
W4384829168 hasLocation W43848291681 @default.
W4384829168 hasLocation W43848291682 @default.
W4384829168 hasLocation W43848291683 @default.
W4384829168 hasOpenAccess W4384829168 @default.
W4384829168 hasPrimaryLocation W43848291681 @default.
W4384829168 hasRelatedWork W2795261237 @default.
W4384829168 hasRelatedWork W3014300295 @default.
W4384829168 hasRelatedWork W3164822677 @default.
W4384829168 hasRelatedWork W4223943233 @default.
W4384829168 hasRelatedWork W4225161397 @default.
W4384829168 hasRelatedWork W4312200629 @default.
W4384829168 hasRelatedWork W4360585206 @default.
W4384829168 hasRelatedWork W4364306694 @default.
W4384829168 hasRelatedWork W4380075502 @default.
W4384829168 hasRelatedWork W4380086463 @default.
W4384829168 hasVolume "23" @default.