conftrace_

gradient clipping

54 papers

Explore in graph

Co-occurring keywords

differential privacy (1016) stochastic gradient descent (1091) neural network optimization (1293) convergence analysis (395) stochastic optimization (1060) non-convex optimization (547) federated learning (1333) heavy-tailed noise (23) heavy-tailed distribution (98) gradient descent (1144)

Papers

Online Convex Optimization with Heavy Tails: Old Algorithms, New Regrets, and Applications ALT 2026
An Improved Privacy and Utility Analysis of Differentially Private SGD with Bounded Domain and Smooth Losses AAAI 2026
ASKD: Reinforcement Learning-Style Knowledge Distillation with Quality-Adaptive Skewness AAAI 2026
Revisiting Gradient Normalization and Clipping for Nonconvex SGD under Heavy-Tailed Noise: Necessity, Sufficiency, and Acceleration JMLR 2025
Dyn-D^2P: Dynamic Differentially Private Decentralized Learning with Provable Utility Guarantee IJCAI 2025
Optimized Gradient Clipping for Noisy Label Learning AAAI 2025
Online Sensitivity Optimization in Differentially Private Learning AAAI 2024
Differentially Private Video Activity Recognition WACV 2024
Near-Optimal Streaming Heavy-Tailed Statistical Estimation with Clipped SGD NIPS 2024
Breaking the Heavy-Tailed Noise Barrier in Stochastic Optimization Problems AISTATS 2024
Efficiently Train ASR Models that Memorize Less and Perform Better with Per-core Clipping INTERSPEECH 2024
Quantifying Unintended Memorization in BEST-RQ ASR Encoders INTERSPEECH 2024
Parameter-free Clipped Gradient Descent Meets Polyak NIPS 2024
Adam-family Methods for Nonsmooth Optimization with Convergence Guarantees JMLR 2024
Byzantine Robustness and Partial Participation Can Be Achieved at Once: Just Clip Gradient Differences NIPS 2024
Effectively Using Public Data in Privacy Preserving Machine Learning ICML 2023
Federated Learning with Client Subsampling, Data Heterogeneity, and Unbounded Smoothness: A New Algorithm and Lower Bounds NIPS 2023
FedNAR: Federated Optimization with Normalized Annealing Regularization NIPS 2023
Revisiting Gradient Clipping: Stochastic bias and tight convergence guarantees ICML 2023
Automatic Clipping: Differentially Private Deep Learning Made Easier and Stronger NIPS 2023
Differentially Private Optimization on Large Model at Small Cost ICML 2023
Improved Convergence in High Probability of Clipped Gradient Methods with Heavy Tailed Noise NIPS 2023
Provably Robust Temporal Difference Learning for Heavy-Tailed Rewards NIPS 2023
Generalized Teacher Forcing for Learning Chaotic Dynamics ICML 2023
A Unified Fast Gradient Clipping Framework for DP-SGD NIPS 2023