[HTML][HTML] Adaptive Kernel Convolutional Stereo Matching Recurrent Network
J Wang, H Sun, P Jia - Sensors, 2024 - mdpi.com
For binocular stereo matching techniques, the most advanced method currently is using an
iterative structure based on GRUs. Methods in this class have shown high performance on …
iterative structure based on GRUs. Methods in this class have shown high performance on …
Kernel normalized convolutional networks for privacy-preserving machine learning
R Nasirigerdeh, J Torkzadehmahani… - … IEEE Conference on …, 2023 - ieeexplore.ieee.org
Normalization is an important but understudied challenge in privacy-related application
domains such as federated learning (FL), differential privacy (DP), and differentially private …
domains such as federated learning (FL), differential privacy (DP), and differentially private …
Efficient Federated and Privacy-preserving Machine Learning
R Nasirigerdeh - 2024 - mediatum.ub.tum.de
We first show that the regression and neural network models trained in a federated fashion
can achieve optimal utility independent of data distribution across clients. Next, we introduce …
can achieve optimal utility independent of data distribution across clients. Next, we introduce …
A Purified Ensemble Framework of Improved Inception Networks for Cervical Cytology Classification
Abstract Background and Objective: Cervical cancer is one of the severe diseases
threatening women's health. Early detection and diagnosis can significantly reduce cancer …
threatening women's health. Early detection and diagnosis can significantly reduce cancer …