InfoNCE Loss
خسارة InfoNCE
دالة خسارة تبايُنية تُعظِّم المعلومات المتبادلة بين الأزواج الإيجابية وتعامل بقية عينات الدفعة كسلبيات.
A contrastive loss function that maximizes mutual information between positive pairs while treating all other in-batch samples as negatives.
Also translated asخسارة التقدير التبايُني للمعلومات، دالة InfoNCE
First appears in this corpus in: Representation Learning with Contrastive Predictive Coding (2018)
Appears in these papers
- Learning Transferable Visual Models from Natural Language Supervision2021in the sky ✦
- Representation Learning with Contrastive Predictive Coding2018in the sky ✦
- ImageBind: One Embedding Space to Bind Them All2023in the sky ✦
- Momentum Contrast for Unsupervised Visual Representation Learning2020in the sky ✦
- A Simple Framework for Contrastive Learning of Visual Representations2020in the sky ✦