Federated Classification in Hyperbolic Spaces via Secure Aggregation of Convex Hulls

Keyphrases

• Convex Hull 100%
• Hyperbolic Space 100%
• Secure Aggregation 100%
• Tree Structure 28%
• Federated Learning 28%
• Phylogeny 14%
• Learning Methods 14%
• Language Processing 14%
• Classification Accuracy 14%
• Small Distortion 14%
• Finite Dimension 14%
• Field Development 14%
• Communication Cost 14%
• Euclidean Space 14%
• Process Graph 14%
• Quantization Method 14%
• Recovery-based 14%
• SVM Classifier 14%
• Client Data 14%
• Client-centered 14%
• Privacy Constraint 14%
• Reed-Solomon 14%
• Label Recovery 14%
• Global Classifier 14%
• Data Leakage 14%
• Diverse Data 14%
• Single-cell RNA-seq Data 14%
• Balanced Graph Partitioning 14%
• Privacy-preserving Machine Learning 14%
• Graph Data Mining 14%
• Label Switching 14%
• Bh Sequences 14%

Mathematics

• Convex Hull 100%
• Hyperbolic Space 100%
• Tree-Like 28%
• Integer 14%
• Euclidean Space 14%
• Theoretic Approach 14%
• Graph Partition 14%
• Finite Dimension 14%
• Euclidean Counterpart 14%
• Initial Step 14%
• Label Switching 14%
• Data Mining 14%

Computer Science

• Hyperbolic Space 100%
• Federated Learning 28%
• Data Mining 14%
• Theoretic Approach 14%
• Classification Accuracy 14%
• Communication Cost 14%
• Privacy Preserving 14%
• Graph Partition 14%
• Support Vector Machine 14%
• Language Processing 14%
• Distributed Version 14%
• Individual Client 14%