Robuta

https://openreview.net/forum?id=iYwiyS1YdQ&referrer=%5Bthe%20profile%20of%20Martin%20Tak%C3%A1%C4%8D%5D(%2Fprofile%3Fid%3D~Martin_Tak%C3%A1%C4%8D1) FedPeWS: Personalized Warmup via Subnetworks for Enhanced Heterogeneous Federated Learning |... Statistical data heterogeneity is a significant barrier to convergence in federated learning (FL). While prior work has advanced heterogeneous FL through... personalizedwarmupviasubnetworksenhanced https://www.uni-augsburg.de/en/vkal/explaining-graph-convolutional-neural-networks-pat Explaining Graph Convolutional Neural Networks: Patient-Specific Subnetworks and Biomarker... convolutional neural networksexplaininggraphpatientspecific https://arxiv.org/abs/2601.22141 [2601.22141] Routing the Lottery: Adaptive Subnetworks for Heterogeneous Data Abstract page for arXiv paper 2601.22141: Routing the Lottery: Adaptive Subnetworks for Heterogeneous Data the lottery2601routingadaptivesubnetworks https://openreview.net/forum?id=fUwfjPzI8g&referrer=%5Bthe%20profile%20of%20Sotirios%20Chatzis%5D(%2Fprofile%3Fid%3D~Sotirios_Chatzis1) Continual Learning via Winning Subnetworks That Arise Through Stochastic Local Competition |... The aim of this work is to address catastrophic forgetting in class-incremental learning. To this end, we propose deep networks that comprise blocks of units... continual learningviawinningsubnetworks https://openreview.net/forum?id=8GZxtu46Kx A Tale of Two Circuits: Grokking as Competition of Sparse and Dense Subnetworks | OpenReview empirical findings on sparsity and subnetwork competition during grokking a tale of two