weight sharing
مشاركة الأوزان
استخدام مجموعة الأوزان نفسها في كل موضع مكاني من الصورة، مما يقلّص المعاملات ويفرض ثبات الإزاحة.
Using the same set of weights at every spatial position, reducing parameters and enforcing translation invariance.
Also translated asتقاسم الأوزان، الأوزان المشتركة
First appears in this corpus in: Signature Verification Using a "Siamese" Time Delay Neural Network (1993)
Appears in these papers
- Unsupervised Visual Representation Learning by Context Prediction2015in the sky ✦
- Gradient-Based Learning Applied to Document Recognition1998in the sky ✦
- Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding2016in the sky ✦
- ELECTRA: Pre-Training Text Encoders as Discriminators Rather Than Generators2020in the sky ✦
- N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting2019in the sky ✦
- N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting2019in the sky ✦
- Perceiver: General Perception with Iterative Attention2021in the sky ✦
- Perceiver: General Perception with Iterative Attention2021in the sky ✦
- RAFT: Recurrent All-Pairs Field Transforms for Optical Flow2020in the sky ✦
- RAFT: Recurrent All-Pairs Field Transforms for Optical Flow2020in the sky ✦
- Recurrent Neural Networks (RNNs): A Gentle Introduction and Overview2019in the sky ✦
- Recurrent Neural Networks (RNNs): A Gentle Introduction and Overview2019in the sky ✦
- Signature Verification Using a "Siamese" Time Delay Neural Network1993in the sky ✦
- Exploring Simple Siamese Representation Learning2021in the sky ✦
- Spectral Networks and Deep Locally Connected Networks on Graphs2014in the sky ✦
- Spectral Networks and Deep Locally Connected Networks on Graphs2014in the sky ✦
- Very Deep Convolutional Networks for Large-Scale Image Recognition2014in the sky ✦