Research map / Optimization / Optimizers
Stochastic Approximate Gradient Descent via the Langevin Algorithm
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 |
|---|---|---|---|
| Fine-tune VAE with SAGD instead of VAE alone | VAE · Normal latent distribution | Kolmogorov-Smirnov distance: 0.033 -> 0.033 | No clear effect |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · Normal latent distribution | 1-Wasserstein distance: 0.052 -> 0.052 | No clear effect |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · Exp(2) latent distribution | Kolmogorov-Smirnov distance: 0.095 -> 0.064 | Helped |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · Exp(2) latent distribution | 1-Wasserstein distance: 0.226 -> 0.115 | Helped |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · Mixture latent distribution | Kolmogorov-Smirnov distance: 0.127 -> 0.085 | Helped |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · Mixture latent distribution | 1-Wasserstein distance: 0.320 -> 0.197 | Helped |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · High-dimensional exponential latent distribution | Kolmogorov-Smirnov distance: 0.093 -> 0.065 | Helped |
| Fine-tune VAE with SAGD instead of VAE alone | VAE · High-dimensional exponential latent distribution | 1-Wasserstein distance: 0.222 -> 0.093 | Helped |
| Use SAGD rather than IWAE for VAE refinement | VAE · Exp(2) latent distribution | 1-Wasserstein distance: 0.125 -> 0.115 | Helped |
| Use SAGD rather than IWAE for VAE refinement | VAE · Mixture latent distribution | Kolmogorov-Smirnov distance: 0.098 -> 0.085 | Helped |
| Use SAGD rather than IWAE for VAE refinement | VAE · Mixture latent distribution | 1-Wasserstein distance: 0.165 -> 0.197 | Hurt |
| Use SAGD rather than IWAE for VAE refinement | VAE · High-dimensional exponential latent distribution | 1-Wasserstein distance: 0.158 -> 0.093 | Helped |
Closest papers by meaning
- A Variational Analysis of Stochastic Gradient Algorithms (ICML 2016 · similarity 0.90)
- Sampling from Gaussian Process Posteriors using Stochastic Gradient Descent (NeurIPS 2023 · similarity 0.90)
- On the Convergence of mSGD and AdaGrad for Stochastic Optimization (ICLR 2022 · similarity 0.90)
- Stochastic Gradient Descent under Markovian Sampling Schemes (ICML 2023 · similarity 0.90)
- Improved Convergence Rate of Stochastic Gradient Langevin Dynamics with Variance Reduction and its Application to Optimization (NeurIPS 2022 · similarity 0.89)
- Statistical Guarantees for High-Dimensional Stochastic Gradient Descent (NeurIPS 2025 · similarity 0.88)
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.