Siamese Network
الشبكة السيامية
بنية تتكوّن من شبكتين عصبيتين متطابقتين تتشاركان جميع الأوزان، يعالج كل منهما مدخلاً مختلفاً ثم تُقارَن مخرجاتهما لقياس التشابه. الفكرة الأساسية هي تعلّم تمثيل تكون فيه المسافة ذات معنى دون الحاجة لتصنيف كل فئة على حدة.
An architecture consisting of two identical neural networks sharing all weights, each processing a different input, with their outputs compared to measure similarity. The core idea is learning a representation where distance is meaningful without needing to classify each category individually.
Also translated asشبكة التوائم، الشبكة المزدوجة
First appears in this corpus in: Signature Verification Using a "Siamese" Time Delay Neural Network (1993)
Appears in these papers
- Barlow Twins: Self-Supervised Learning via Redundancy Reduction2021in the sky ✦
- FaceNet: A Unified Embedding for Face Recognition and Clustering2015in the sky ✦
- Sentence-BERT: Sentence Embeddings Using Siamese BERT-Networks2019in the sky ✦
- Sentence-BERT: Sentence Embeddings Using Siamese BERT-Networks2019in the sky ✦
- Signature Verification Using a "Siamese" Time Delay Neural Network1993in the sky ✦
- Exploring Simple Siamese Representation Learning2021in the sky ✦
- Exploring Simple Siamese Representation Learning2021in the sky ✦
- VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning2021in the sky ✦