Tag: domain shift

A systematic review of 120 papers finds deep learning DDoS detectors excel in experiments but struggle in real-world cloud environments, ...

A hybrid deep learning framework for classifying cataract, diabetic retinopathy, glaucoma, and normal retinas dropped from 99.24 percent internal accuracy ...

A new study shows that a frozen CLIP vision-language model with lightweight adapters can improve human activity recognition in low-light ...

A new framework called TRRN re-indexes and filters web-retrieved images against few-shot class anchors, lifting accuracy across six benchmark settings ...

A CORAL-based domain adaptation framework significantly improved machine learning predictions of student graduation when models were transferred between academic programs ...

A new test-time adaptation framework shows that online model updates can rescue surgical instrument segmentation from severe low light and ...

A new entropy-guided framework prunes up to 95 percent of Wi-Fi access points from indoor positioning fingerprints while preserving accuracy ...

A new review of 29 recent studies finds that no deep learning intrusion detection system has been evaluated jointly for ...

Researchers combined diffusion-generated synthetic retinal images, contrastive self-supervised pre-training, and a five-model ensemble to reach 95.1 percent accuracy in diabetic ...

Researchers have built a hybrid transformer-U-Net that segments diabetic foot ulcers with expert-level overlap and quantitatively explainable visualisations.