Byzantine-resilient federated learning leveraging confidence score to identify retinal disease

Keyphrases

• Resilient 100%
• Retinal Diseases 100%
• Byzantine Empire 100%
• VGG19 100%
• Federated Learning 100%
• Confidence Score 100%
• InceptionV3 100%
• Global Model 66%
• F1 Score 66%
• Retinal OCT Image 66%
• Healthcare 33%
• Client-centered 33%
• Deep Learning 33%
• Area under the Receiver Operating Characteristic Curve 33%
• Medical Image Processing 33%
• Process Data 33%
• Precision-recall 33%
• Data Privacy 33%
• Privacy Concerns 33%
• ROC-AUC 33%
• Common Model 33%
• Multiple Actors 33%
• Diabetic Macular Edema 33%
• Local Data 33%
• Age-related Macular Degeneration 33%
• Data Poisoning 33%
• Distributed Machine Learning 33%
• Model Precision 33%
• Poisoning Attack 33%
• Local Update 33%
• Malicious Clients 33%
• Machine Learning Paradigms 33%
• Sensitive Domains 33%

Computer Science

• Federated Learning 100%
• VGG-19 Convolutional Neural Network 100%
• Inception V3 100%
• Confidence Score 100%
• Medical Image Processing 33%
• Deep Learning Method 33%
• Data Privacy 33%
• Characteristic Curve 33%
• Distributed Machine Learning 33%
• Malicious Client 33%
• Sensitive Domain 33%
• Processing Data 33%