gradient noise

Beyond Sharpness: The Role of Nonuniformity in Generalization AAAI 2026 Limitations of SGD for Multi-Index Models Beyond Statistical Queries COLT 2026 Adaptive Methods through the Lens of SDEs: Theoretical Insights on the Role of Noise ICLR 2025 Bayesian Weight Enhancement with Steady-State Adaptation for Test-time Adaptation in Dynamic Environments ICML 2025 Formation of Representations in Neural Networks ICLR 2025 Subspace Optimization for Large Language Models with Convergence Guarantees ICML 2025 Accelerated Convergence of Stochastic Heavy Ball Method under Anisotropic Gradient Noise ICLR 2024 Dealing With Unbounded Gradients in Stochastic Saddle-point Optimization ICML 2024 DOPPLER: Differentially Private Optimizers with Low-pass Filter for Privacy Noise Reduction NeurIPS 2024 EAT: Towards Long-Tailed Out-of-Distribution Detection AAAI 2024 Parameter Symmetry and Noise Equilibrium of Stochastic Gradient Descent NeurIPS 2024 Robust Approximate Sampling via Stochastic Gradient Barker Dynamics AISTATS 2024 Stochastic Approximation with Decision-Dependent Distributions: Asymptotic Normality and Optimality JMLR 2024 SWAP: Sparse Entropic Wasserstein Regression for Robust Network Pruning ICLR 2024 Symmetry Induces Structure and Constraint of Learning ICML 2024 Large deviations rates for stochastic gradient descent with strongly convex functions AISTATS 2023 On the Overlooked Structure of Stochastic Gradients NeurIPS 2023 Stochastic Collapse: How Gradient Noise Attracts SGD Dynamics Towards Simpler Subnetworks NeurIPS 2023 STGN: an Implicit Regularization Method for Learning with Noisy Labels in Natural Language Processing EMNLP 2022 Adaptive Privacy Preserving Deep Learning Algorithms for Medical Data WACV 2021 Differentiable Annealed Importance Sampling and the Perils of Gradient Noise NeurIPS 2021 Instance Enhancement Batch Normalization: An Adaptive Regulator of Batch Noise AAAI 2020 On the Noisy Gradient Descent that Generalizes as SGD ICML 2020 Robustness Analysis of Non-Convex Stochastic Gradient Descent using Biased Expectations NeurIPS 2020 On Acceleration with Noise-Corrupted Gradients ICML 2018