Research map / Probabilistic and scientific modeling / Variational and generative sampling
Meta-Learning For Stochastic Gradient MCMC
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 the meta-learned sampler instead of SGHMC | Bayesian MLP classification · MNIST architecture generalization (NT) | test Err: 98.21$\pm0.01$% -> 98.36$\pm \pmb{0.02}$% | Helped |
| use the meta-learned sampler instead of SGHMC | Bayesian MLP classification · MNIST architecture generalization (NT) | test NLL: 705$\pm 3.44$ -> 640$\pm 6.25$ | Helped |
| use the meta-learned sampler instead of SGHMC | Bayesian MLP classification · MNIST activation generalization (AF) | test NLL: 929$\pm 2.95$ -> \textbf{875$\pm\pmb{3.19}$} | Helped |
| use the meta-learned sampler instead of SGHMC | Bayesian MLP classification · MNIST dataset generalization (Data) | test NLL: 246$\pm 5.43$ -> 230$\pm 3.23$ | Helped |
| use the meta-learned sampler instead of SGHMC | Bayesian MLP classification · MNIST activation generalization (AF) | test Err: 97.72$\pm \pmb{0.01}$% -> 97.72$\pm \pmb{0.02}$% | No clear effect |
| use the meta-learned sampler instead of SGHMC | Bayesian MLP classification · MNIST dataset generalization (Data) | test Err: 98.52$\pm 0.03$% -> 98.62$\pm \pmb{0.02}$% | Helped |
| use the meta-learned sampler instead of SGHMC | SG-MCMC for noisy-gradient Gaussian sampling · 20D Gaussian test | effective sample size: 22 -> 59 | Helped |
| remove the offset β from fφQ (ablation) | NNSGHMC · JSB chorales | test NLL | Helped |
| remove the offset β from fφQ (ablation) | NNSGHMC · JSB chorales | convergence speed | No clear effect |
Closest papers by meaning
- High-Order Stochastic Gradient Thermostats for Bayesian Learning of Deep Models (AAAI 2016 · similarity 0.92)
- Learning to Explore for Stochastic Gradient MCMC (ICML 2024 · similarity 0.91)
- Structured Stochastic Gradient MCMC (ICML 2022 · similarity 0.91)
- Parameter Expanded Stochastic Gradient Markov Chain Monte Carlo (ICLR 2025 · similarity 0.91)
- A Hybrid Stochastic Gradient Hamiltonian Monte Carlo Method (AAAI 2021 · similarity 0.90)
- Stochastic Gradient Richardson-Romberg Markov Chain Monte Carlo (NeurIPS 2016 · similarity 0.90)
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.