RELTA-SGLD: Relative-Growth Localized Taming for Nonconvex Stochastic-Gradient Langevin Learning

arXiv:2607.19544 · stat.ML, cs.LG, math.OC · Submitted 2026-07-21 · Read on arXiv

Yiwei Zhou, Ziheng Chen

stat.ML, cs.LG, math.OC

Submitted: 2026-07-21

Comments: 40 pages, 1 figure, 4 tables, Code available at: https://github.com/M381C/RELTA

Code: https://github.com/M381C/RELTA

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Accuracy — 23× in this paper · explained in 1 episode
• Langevin Dynamics — 8× in this paper · explained in 3 episodes
• Absorb — 5× in this paper · explained in 1 episode
• Regularization — 5× in this paper · explained in 3 episodes
• Verification — 4× in this paper · explained in 2 episodes
• Generator — 3× in this paper · explained in 2 episodes
• Monotonicity — 3× in this paper · explained in 2 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Active Learning — 1× in this paper · explained in 6 episodes
• Calibration — 1× in this paper · explained in 9 episodes
• Decoupling — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 14 episodes
• Hilbert Space — 1× in this paper · explained in 2 episodes
• Hyperparameters — 1× in this paper · explained in 2 episodes
• Optimal Transport — 1× in this paper · explained in 5 episodes
• Wasserstein Distance — 1× in this paper · explained in 6 episodes
• Weight Decay — 1× in this paper · explained in 2 episodes

Sources

• Deterministic Envelopes for Tamed SGLD: Decoupling Stochastic Gradient Noise and Localizing Taming
• Deterministic Denominator Design for Localized Tamed Stochastic Gradient Langevin Dynamics
• kTULA: A Langevin sampling algorithm with improved KL bounds under super-linear log-gradients
• A modified tamed scheme for stochastic differential equations with superlinear drifts
• Error estimates for tamed Euler and Randomized Euler schemes for SDEs with locally Lipschitz drift with applications to non-logconcave sampling and optimization

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• Scalable Posterior Uncertainty for Flexible Density-Based Clustering
• A Transport Geometry of Benign Overfitting
• Efficient quadratic entropy with distance sketches