Please join us in gather.town (see link above). To see the abstracts of the posters presented in this session, please see below the schedule.

Authors/papers presenting posters in gather.town for this session:

• Gaussian Graphical Models as an Ensemble Method for Distributed Gaussian Processes, Hamed Jalali

• DAdaQuant: Doubly-adaptive quantization for communication-efficient Federated Learning, Robert Hönig

• Using a one dimensional parabolic model of the full-batch loss to estimate learning rates during training, Maximus Mutschler

• COCO Denoiser: Using Co-Coercivity for Variance Reduction in Stochastic Convex Optimization, Manuel Madeira

• Decentralized Personalized Federated Learning: Lower Bounds and Optimal Algorithm for All Personalization Modes, Abdurakhmon Sadiev

• Shifted Compression Framework: Generalizations and Improvements, Egor Shulgin

• Faking Interpolation Until You Make It, Alasdair J Paren

• Towards Modeling and Resolving Singular Parameter Spaces using Stratifolds, Pascal M Esser

• Spherical Perspective on Learning with Normalization Layers, Simon W Roburin

• Adaptive Optimization with Examplewise Gradients, Julius Kunze

• On the Relation between Distributionally Robust Optimization and Data Curation, Agnieszka Słowik

• Fast, Exact Subsampled Natural Gradients and First-Order KFAC, Frederik Benzing

• Understanding Memorization from the Perspective of Optimization via Efficient Influence Estimation, Futong Liu

• Community-based Layerwise Distributed Training of Graph Convolutional Networks, Hongyi Li

• A New Scheme for Boosting with an Avarage Margin Distribution Oracle, Ryotaro Mitsuboshi

• Better Linear Rates for SGD with Data Shuffling, Grigory Malinovsky

• Structured Low-Rank Tensor Learning, Jayadev Naram

• ANITA: An Optimal Loopless Accelerated Variance-Reduced Gradient Method, Zhize Li

• EF21 with Bells & Whistles: Practical Algorithmic Extensions of Modern Error Feedback, Igor Sokolov

• On Server-Side Stepsizes in Federated Optimization: Theory Explaining the Heuristics, Grigory Malinovsky