method · first seen 5 Oct 2026
Langevin dynamics
Sampling method that follows noisy gradient flow of the log-density to draw samples from a target distribution.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Sharp dimensional analysis of midpoint methods for Langevin samplingPapers · 5 Oct 2026 · Fan Chen, Sinho Chewi, Jianfeng Lu, Matthew S. Zhang
- Diffusion Removes Langevin's Conditioning Dependence: A Sharp Gaussian AnalysisPapers · 8 Oct 2026 · Adam Perbost, Francis Bach, Pierre Marion
- Global Convergence in Deep Networks via the Second Law of ThermodynamicsPapers · 17 Oct 2026 · Matan Tsipory, Ofir Gaash, Itamar Harel, Nathan Srebro, Daniel Soudry
- Averaged Annealed Markov Chains Achieve the Algorithmic Threshold for Tensor PCA, Part II: Langevin DynamicsPapers · 28 Oct 2026 · Aniket Das, Tselil Schramm
- Zeroth-Order Langevin Dynamics for Stochastic Nonconvex Optimization: Curvature Regularization and Non-Asymptotic GuaranteesPapers · 7 Nov 2026 · Stefano Bruno, Dong-Young Lim
Latest sources
- Zeroth-Order Langevin Dynamics for Stochastic Nonconvex Optimization: Curvature Regularization and Non-Asymptotic GuaranteesPapers · 7 Nov 2026 · Stefano Bruno, Dong-Young Lim
- Averaged Annealed Markov Chains Achieve the Algorithmic Threshold for Tensor PCA, Part II: Langevin DynamicsPapers · 28 Oct 2026 · Aniket Das, Tselil Schramm
- Global Convergence in Deep Networks via the Second Law of ThermodynamicsPapers · 17 Oct 2026 · Matan Tsipory, Ofir Gaash, Itamar Harel, Nathan Srebro, Daniel Soudry
- Diffusion Removes Langevin's Conditioning Dependence: A Sharp Gaussian AnalysisPapers · 8 Oct 2026 · Adam Perbost, Francis Bach, Pierre Marion
- Sharp dimensional analysis of midpoint methods for Langevin samplingPapers · 5 Oct 2026 · Fan Chen, Sinho Chewi, Jianfeng Lu, Matthew S. Zhang