Sebastian U Stich

Monotone Near-Zero-Sum Games: A Generalization of Convex-Concave Minimax

Composite Optimization with Error Feedback: the Dual Averaging Approach

Non-Convex Federated Optimization under Cost-Aware Client Selection

Sequential Subspace Noise Injection Prevents Accuracy Collapse in Certified Unlearning

On the Stability of Nonlinear Dynamics in GD and SGD: Beyond Quadratic Potentials

Revisiting the Effect of Topologies on Decentralized SGD

DADA: Dual Averaging with Distance Adaptation

Collaborative Unpaired Multimodal Learning for Image Classification

FedMuon: Federated Learning with Bias-corrected LMO-based Optimization

Revisiting Consensus Error: A Fine-grained Analysis of Local SGD under Second-order Data Heterogeneity

Towards Faster Decentralized Stochastic Optimization with Communication Compression

Decoupled SGDA for Games with Intermittent Strategy Communication

Optimizing $(L_0, L_1)$-Smooth Functions by Gradient Methods

Exploiting Similarity for Computation and Communication-Efficient Decentralized Optimization

Scalable Decentralized Learning with Teleportation

Decoupled SGDA for Games with Intermittent Strategy Communication