Research map / Probabilistic and scientific modeling / Variational and generative sampling
Aggregated Gradient Langevin Dynamics
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 variance-reduced aggregated gradients instead of full-gradient LMC or stochastic-gradient SGLD | SVRG-LD, SAGA-LD, and TMU-RA · Gaussian mixture distribution | qualitative approximation to target distribution | Helped |
| Use TMU snapshot updating instead of PPU or PTU | AGLD · Gaussian mixture distribution | qualitative approximation to target distribution | Helped |
| Use TMU instead of PPU or PTU | AGLD with the same data-access strategy · YearPredictionMSD test split | test MSE | Helped |
| Use SVRG+ or PPU instead of LMC, SGLD, or PTU | Bayesian ridge regression · YearPredictionMSD test split | test MSE | Helped |
| Use random access instead of cyclic access or random reshuffle | TMU · Bayesian ridge regression, YearPredictionMSD | test MSE | Helped |
| Use cyclic access instead of random access | AGLD methods · Bayesian logistic regression, criteo and kdd12 | time efficiency | Helped |
| Use cyclic access rather than random access as available physical memory decreases | AGLD methods · Bayesian logistic regression, criteo and kdd12 | time efficiency | Helped |
| Use TMU instead of other snapshot-update strategies | AGLD with the same data-access strategy · Bayesian logistic regression, criteo and kdd12 | test log-likelihood | Helped |
| Use TMU-RA instead of TMU-CA or TMU-RR when memory is sufficient | TMU · Bayesian ridge regression | test MSE | Helped |
| Use TMU-CA instead of random-access TMU when memory is insufficient | TMU · Bayesian logistic regression, criteo and kdd12 | time efficiency | Helped |
Closest papers by meaning
- Stochastic Gradient Richardson-Romberg Markov Chain Monte Carlo (NeurIPS 2016 · similarity 0.92)
- The promises and pitfalls of Stochastic Gradient Langevin Dynamics (NeurIPS 2018 · similarity 0.90)
- A Dynamical System View of Langevin-Based Non-Convex Sampling (NeurIPS 2023 · similarity 0.90)
- Stochastic Quasi-Newton Langevin Monte Carlo (ICML 2016 · similarity 0.90)
- A Contour Stochastic Gradient Langevin Dynamics Algorithm for Simulations of Multi-modal Distributions (NeurIPS 2020 · similarity 0.89)
- Sequential Controlled Langevin Diffusions (ICLR 2025 · similarity 0.89)
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.