Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• The model that achieves more class-balanced sampling, especially for minority classes, consistently leads to better final performance in federated active learning.result 90%
• Global-model querying is beneficial only when the global distribution is highly imbalanced and client data are relatively homogeneous; otherwise, the local model is preferable.result 90%
• FairFAL outperforms state-of-the-art approaches on five benchmarks under challenging long-tailed and non-IID settings.result 90%
• FairFAL infers global imbalance and local-global divergence via lightweight prediction discrepancy to enable adaptive selection between global and local query models.method 90%
• FairFAL performs prototype-guided pseudo-labeling using global features to promote class-aware querying.method 90%
• FairFAL applies a two-stage uncertainty-diversity balanced sampling strategy with k-center refinement.method 90%
• Federated active learning effectiveness degrades in realistic settings with severe global class imbalance and highly heterogeneous clients.limitation 85%
• FairFAL is an adaptive class-fair federated active learning framework designed to address extreme non-IID and global class imbalance.method 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card