method · first seen 9 Oct 2026
stochastic gradient Langevin Monte Carlo
Also: SGLD, SG-LMC, stochastic gradient Langevin dynamics
A sampling method combining stochastic gradient descent with Langevin dynamics to draw samples from posterior distributions.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Robust Langevin Sampling under State-Dependent Heavy-Tailed Noise with Applications to LLM Fine-TuningPapers · 9 Oct 2026 · M. Berk Sahin, Ege C. Kaya, Arda Fazla, Behzad Sharif, Abolfazl Hashemi
- Generalization in Learning: Eyring Formula and Predator-prey ModelPapers · 25 Oct 2026 · S. V. Kozyrev, I. A. Lopatin, Alexander Pechen
- Statistical and Privacy Guarantees for Deep M-Regression via Stochastic Gradient Langevin DynamicsPapers · 5 Nov 2026 · Yeheng Ge, Jian Huang, Shuyuan Wu
Latest sources
- Statistical and Privacy Guarantees for Deep M-Regression via Stochastic Gradient Langevin DynamicsPapers · 5 Nov 2026 · Yeheng Ge, Jian Huang, Shuyuan Wu
- Generalization in Learning: Eyring Formula and Predator-prey ModelPapers · 25 Oct 2026 · S. V. Kozyrev, I. A. Lopatin, Alexander Pechen
- Robust Langevin Sampling under State-Dependent Heavy-Tailed Noise with Applications to LLM Fine-TuningPapers · 9 Oct 2026 · M. Berk Sahin, Ege C. Kaya, Arda Fazla, Behzad Sharif, Abolfazl Hashemi