Ahmed Radwan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ahmed Radwan has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Mohamed Sami Shehata
The University of British Columbia
5 shared publications - Omar Abdelaziz
The University of British Columbia
2 shared publications - Ahmad Abdel-Qader
The University of British Columbia
1 shared publication Ahmed Radwan
1 shared publication- Charlene Ronquillo
University of British Columbia, Vancouver, Canada
1 shared publication - Islam Osman
The University of British Columbia
1 shared publication
Latest Publications
- 9 September 2026
- Published by Springer Nature
- p. 212-228
Variational Patch Gating for Training-Free Few-Shot Classification
- 9 September 2026
- Published by Springer Nature
- p. 212-228
Variational Patch Gating for Training-Free Few-Shot Classification
- 22 February 2026
- Published by Springer Nature in AI and Ethics
- Vol. 6 (2) , 165
Secondary use of health data: applications, models, algorithms, and ethical considerations
- 22 February 2026
- Published by Springer Nature in AI and Ethics
- Vol. 6 (2) , 165
Secondary use of health data: applications, models, algorithms, and ethical considerations
- Open Access
GNN-MoE: Context-Aware Patch Routing using GNNs for Parameter-Efficient Domain Generalization
Domain generalization (DG) seeks robust Vision Transformer (ViT) performance on unseen domains. Efficiently adapting pretrained ViTs for DG is challenging; standard fine-tuning is costly and can impair generalization. We propose GNN-MoE, enhancing Parameter-Efficient Fine-Tuning (PEFT) for DG with a Mixture-of-Experts (MoE) framework using efficient Kronecker adapters. Instead of token-based routing, a novel Graph Neural Network (GNN) router (GCN, GAT, SAGE) operates on inter-patch graphs to dynamically assign patches to specialized experts. This context-aware GNN routing leverages inter-patch relationships for better adaptation to domain shifts. GNN-MoE achieves state-of-the-art or competitive DG benchmark performance with high parameter efficiency, highlighting the utility of graph-based contextual routing for robust, lightweight DG.
Open AccessGNN-MoE: Context-Aware Patch Routing using GNNs for Parameter-Efficient Domain Generalization
Domain generalization (DG) seeks robust Vision Transformer (ViT) performance on unseen domains. Efficiently adapting pretrained ViTs for DG is challenging; standard fine-tuning is costly and can impair generalization. We propose GNN-MoE, enhancing Parameter-Efficient Fine-Tuning (PEFT) for DG with a Mixture-of-Experts (MoE) framework using efficient Kronecker adapters. Instead of token-based routing, a novel Graph Neural Network (GNN) router (GCN, GAT, SAGE) operates on inter-patch graphs to dynamically assign patches to specialized experts. This context-aware GNN routing leverages inter-patch relationships for better adaptation to domain shifts. GNN-MoE achieves state-of-the-art or competitive DG benchmark performance with high parameter efficiency, highlighting the utility of graph-based contextual routing for robust, lightweight DG.
- 12 May 2025
- Published by Springer Nature
- p. 100-115
TF-OCM: Training-Free Optimal Community Matching for Domain Generalized Few-Shot Learning
- 12 May 2025
- Published by Springer Nature
- p. 100-115
TF-OCM: Training-Free Optimal Community Matching for Domain Generalized Few-Shot Learning
- 10 March 2025
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 28 (2) , 1-15
Fedpartwhole: federated domain generalization via consistent part-whole hierarchies
- 10 March 2025
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 28 (2) , 1-15
Fedpartwhole: federated domain generalization via consistent part-whole hierarchies