Invariance
الثبات
خاصية لا يتغير فيها ناتج الدالة عند خضوع المدخل لتحويلات معينة كالدوران والانسحاب، وهي الخاصية الجوهرية لانتباه النقاط الثابتة في ألفافولد.
A property where the output of a function does not change when the input undergoes certain transformations such as rotation and translation — the key property of AlphaFold's Invariant Point Attention.
Also translated asاللاتغايُر، عدم التغيّر
First appears in this corpus in: Spatial Transformer Networks (2015)
Appears in these papers
- Highly Accurate Protein Structure Prediction with AlphaFold2021in the sky ✦
- Barlow Twins: Self-Supervised Learning via Redundancy Reduction2021in the sky ✦
- Barlow Twins: Self-Supervised Learning via Redundancy Reduction2021in the sky ✦
- Dynamic Routing Between Capsules2017in the sky ✦
- Exploring Simple Siamese Representation Learning2021in the sky ✦
- Exploring Simple Siamese Representation Learning2021in the sky ✦
- Spatial Transformer Networks2015in the sky ✦
- VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning2021in the sky ✦