data heterogeneity
The variation and distribution differences across datasets, which can affect model performance and generalization, particularly in distributed learning scenarios where data may be non-IID (independent and identically distributed).
- Covariances for Free: Exploiting Mean Distributions for Training-free Federated Learning
- DKDR: Dynamic Knowledge Distillation for Reliability in Federated Learning
- Diffusion Federated Dataset
- Efficient Federated Learning against Byzantine Attacks and Data Heterogeneity via Aggregating Normalized Gradients
- Error Feedback under $(L_0,L_1)$-Smoothness: Normalization and Momentum
- Exact and Linear Convergence for Federated Learning under Arbitrary Client Participation is Attainable
- FedEL: Federated Elastic Learning for Heterogeneous Devices
- FedFACT: A Provable Framework for Controllable Group-Fairness Calibration in Federated Learning
- FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning
- FedLPA: Local Prior Alignment for Heterogeneous Federated Generalized Category Discovery
- FedRTS: Federated Robust Pruning via Combinatorial Thompson Sampling
- Global Prompt Refinement with Non-Interfering Attention Masking for One-Shot Federated Learning
- Learning to Route: Per-Sample Adaptive Routing for Multimodal Multitask Prediction
- Learning to Specialize: Joint Gating-Expert Training for Adaptive MoEs in Decentralized Settings
- Personalized Federated Conformal Prediction with Localization
- PubSub-VFL: Towards Efficient Two-Party Split Learning in Heterogeneous Environments via Publisher/Subscriber Architecture
- Ravan: Multi-Head Low-Rank Adaptation for Federated Fine-Tuning
- Revisiting Consensus Error: A Fine-grained Analysis of Local SGD under Second-order Data Heterogeneity
- Second-order Optimization under Heavy-Tailed Noise: Hessian Clipping and Sample Complexity Limits
- Soft-consensual Federated Learning for Data Heterogeneity via Multiple Paths
- Tackling Feature-Classifier Mismatch in Federated Learning via Prompt-Driven Feature Transformation
- Transfer Learning for Benign Overfitting in High-Dimensional Linear Regression
- Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training