FedOUI downweights federated clients whose activation patterns on a probe batch deviate from the round-wise OUI distribution, yielding gains under strong non-IID and noisy conditions on CIFAR-10.
and Dolz, Manuel F
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.LG 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
OUIDecay adaptively rescales layer-wise weight decay in CNNs using an online activation-based Overfitting-Underfitting Indicator and outperforms fixed decay in 7 of 8 tested settings.
OUI provides an activation-based observable that anticipates training regimes across supervised learning, reinforcement learning, and control tasks before convergence occurs.
citing papers explorer
-
FedOUI: OUI-Guided Client Weighting for Federated Aggregation
FedOUI downweights federated clients whose activation patterns on a probe batch deviate from the round-wise OUI distribution, yielding gains under strong non-IID and noisy conditions on CIFAR-10.
-
OUIDecay: Adaptive Layer-wise Weight Decay for CNNs Using Online Activation Patterns
OUIDecay adaptively rescales layer-wise weight decay in CNNs using an online activation-based Overfitting-Underfitting Indicator and outperforms fixed decay in 7 of 8 tested settings.
-
OUI as a Structural Observable: Towards an Activation-Centric View of Neural Network Training
OUI provides an activation-based observable that anticipates training regimes across supervised learning, reinforcement learning, and control tasks before convergence occurs.