FedCova: Robust Federated Covariance Learning Against Noisy Labels
Noisy labels in distributed datasets induce severe local overfitting and consequently compromise the global model in federated learning (FL).
Academic or research source. Check the methodology, sample size, and whether it's been replicated.
Key Takeaways
- Potential technical breakthrough.
- Noisy labels in distributed datasets induce severe local overfitting and consequently compromise the global model in federated learning (FL).
- Most existing solutions rely on selecting clean devices or aligning with public clean datasets, rather than endowing the model itself with robustness.
What It Means
Context
Noisy labels in distributed datasets induce severe local overfitting and consequently compromise the global model in federated learning (FL). Most existing solutions rely on selecting clean devices or aligning with public clean datasets, rather than endowing the model itself with robustness. In this paper, we propose FedCova, a dependency-free federated covariance learning framework that eliminates such external reliances by enhancing the model's intrinsic robustness via a new perspective on feature covariances. Specifically, FedCova encodes data into a discriminative but resilient feature space to tolerate label noise. Built on mutual information maximization, we design a novel objective for federated lossy feature encoding that relies solely on class feature covariances with an error tolerance term. Leveraging feature subspaces characterized by covariances, we construct a subspace-augmented federated classifier. FedCova unifies three key processes through the covariance: (1) training the network for feature encoding, (2) constructing a classifier directly from the learned features, and (3) correcting noisy labels based on feature subspaces. We implement FedCova across both…
For builders
Most existing solutions rely on selecting clean devices or aligning with public clean datasets, rather than endowing the model itself with robustness.
For Builders
Most existing solutions rely on selecting clean devices or aligning with public clean datasets, rather than endowing the model itself with robustness.