Statistical Learning
Protected Test-Time Adaptation via Online Entropy Matching: A Betting Approach Y arin Bar
We present a novel approach for test-time adaptation via online self-training, consisting of two components.
SHOT: Suppressing the Hessian along the Optimization Trajectory for Gradient-Based Meta-Learning
To validate the effectiveness of SHOT, we conduct empirical tests on standard few-shot learning tasks and qualitatively analyze its dynamics.
Federated Learning over Connected Modes Dennis Grinwald
In each communication round, each client uniformly samples points in the neighborhood of their projected point (top-right plot), and jointly train the solution simplex.
Performance Scaling via Optimal Transport: Enabling Data Selection from Partially Revealed Sources
Optimal Transport distance to predict the model's performance for any data mixture
Towards Understanding the Dynamics of Gaussian-Stein Variational Gradient Descent
Stein V ariational Gradient Descent (SVGD) is a nonparametric particle-based deterministic sampling algorithm.