Research map / Optimization / Optimizers
Sampling with Trusthworthy Constraints: A Variational Gradient Framework
Xingchao Liu, Xin Tong, qiang liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use constraint-controlled gradient (Control) instead of primal-dual updates of the multiplier | Langevin and SVGD constrained sampling · Loan Application logic constraint | logic loss (train and test) | Improved |
| use SVGD instead of Langevin as the base sampler | constrained sampling methods · Loan Application logic constraint | test accuracy | Improved |
| use constraint-controlled gradient instead of primal-dual | constrained sampling methods · COMPAS monotonic Bayesian neural network | monotonic loss vs log-likelihood trade-off | Improved |
| use constraint-controlled gradient instead of primal-dual | constrained sampling methods · Adult Income fair Bayesian neural network | constraint enforcement speed and strictness | Improved |
| add the constraint to NCSN Langevin sampling using Control+Langevin | NCSN on CIFAR10 · black square center constraint, epsilon 0.0001 | constraint satisfaction and sample diversity | Improved |
Closest papers by meaning
- Functional Gradient Flows for Constrained Sampling (NeurIPS 2024 · similarity 0.91)
- Penalized Overdamped and Underdamped Langevin Monte Carlo Algorithms for Constrained Sampling (JMLR 2024 · similarity 0.90)
- Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm (NeurIPS 2016 · similarity 0.90)
- Constrained Sampling with Primal-Dual Langevin Monte Carlo (NeurIPS 2024 · similarity 0.90)
- Sampling in Constrained Domains with Orthogonal-Space Variational Gradient Descent (NeurIPS 2022 · similarity 0.90)
- Fast Non-Log-Concave Sampling under Nonconvex Equality and Inequality Constraints with Landing (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
