Interested in training hyperspectral image analysis models with reduced annotated data?
Salma explores this question in her recent paper.
doi.org/10.1007/s104...
#hyperspectral #HSI #AI #ML #ContrastiveLearning
#sqIRL #IDLab #UAntwerp #imec
Latest posts tagged with #ContrastiveLearning on Bluesky
Interested in training hyperspectral image analysis models with reduced annotated data?
Salma explores this question in her recent paper.
doi.org/10.1007/s104...
#hyperspectral #HSI #AI #ML #ContrastiveLearning
#sqIRL #IDLab #UAntwerp #imec
Semantic Retrieval Contrastive Learning Boosts Recommendations
A new semantic retrieval contrastive learning approach claims to boost recommendation accuracy, according to the latest article. Read more: getnews.me/semantic-retrieval-contr... #semanticretrieval #contrastivelearning #recommendations
Task-Level Contrastiveness Boosts Cross-Domain Few-Shot Learning
A new task-level contrastive method boosts few-shot learning accuracy on the MetaDataset benchmark without adding parameters or extra computation. Read more: getnews.me/task-level-contrastivene... #fewshotlearning #contrastivelearning #metadataset
Semantic Retrieval Contrastive Learning (SRA-CL) Improves Recommendations
SRA‑CL mixes large language models with contrastive learning to produce richer embeddings, improving recommendations. Tests on four benchmark datasets show gains. getnews.me/semantic-retrieval-contr... #semanticretrieval #contrastivelearning
GenView++ Enhances Contrastive Learning with Adaptive Views
GenView++ improves MoCo v2 on ImageNet linear classification by +2.5% and raises zero‑shot accuracy by +12.31% over CLIP across ten benchmarks overall. Read more: getnews.me/genview-enhances-contras... #genview #contrastivelearning
Spatially Regularized Contrastive Learning Improves Geo‑Localization
Semivariogram‑based spatial regularization improves GeoCLIP accuracy on the OSV5M street‑view dataset, helping it separate visually similar but distant locations. Read more: getnews.me/spatially-regularized-co... #geolocalization #contrastivelearning
Diffusion‑Augmented Contrastive Learning: Robust Biosignal Encoding
Diffusion‑Augmented Contrastive Learning (DACL) achieved an AUROC of 0.7815 on the PhysioNet 2017 ECG benchmark, showing robust biosignal encoding. Read more: getnews.me/diffusion-augmented-cont... #biosignals #diffusion #contrastivelearning
Contrastive Learning Boosts Machine Unlearning Effectiveness
CoUn uses contrastive learning on retained data to push forgotten samples away in latent space, boosting machine‑unlearning effectiveness without full retraining. Read more: getnews.me/contrastive-learning-boo... #machineunlearning #contrastivelearning #ai
Contrastive Learning Boosts Instrument and Synthesizer Retrieval
A contrastive learning model retrieves instrument timbres from isolated or mixed audio, reaching 81.7% top‑1 and 95.7% top‑5 accuracy on three‑instrument queries. Read more: getnews.me/contrastive-learning-boo... #contrastivelearning #timbre
👉 Let’s talk multimodal learning — come by or check out the paper!
Joint work with an amazing team: @benoit-dufumier.bsky.social, D. Tuia & J-P. Thiran
#ICLR2025 #MultimodalLearning #SelfSupervisedLearning #ContrastiveLearning
"NYCerebro is a CLIP-powered search engine for NYC traffic cameras. It uses AI to find camera views matching your text descriptions."
nycerebro.vercel.app
#solidstatelife #ai #contrastivelearning