Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Spectral Approach to Gradient Estimation for Implicit Distributions
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 SSGE instead of Stein+ for gradient estimates | Gradient-free HMC · UCI Glass hyperparameter posterior | average acceptance ratio | Helped |
| Use SSGE instead of KMC for gradient-free HMC | Gradient-free HMC · UCI Glass hyperparameter posterior | average acceptance ratio | Helped |
| Use SSGE for out-of-sample gradient estimation | Gradient estimation for a standard Gaussian · 1-D standard Gaussian | gradient estimation accuracy | Helped |
| Add the entropy term, estimated using SSGE (ablation) | Implicit VAE · CelebA | sample quality | Helped |
| Add SSGE-estimated entropy gradients to the Implicit VAE | Implicit VAE · CelebA | FID | Helped |
| Use SSGE with an implicit posterior instead of a factorized Gaussian posterior | Bayesian neural network · 1-D regression | predictive uncertainty | Mixed |
| Use SSGE with a factorized posterior instead of BBB | Bayesian neural network · 1-D regression | prediction | No clear effect |
Closest papers by meaning
- Gradient Estimation with Discrete Stein Operators (NeurIPS 2022 · similarity 0.89)
- Kernel Semi-Implicit Variational Inference (ICML 2024 · similarity 0.88)
- Sticking the Landing: Simple, Lower-Variance Gradient Estimators for Variational Inference (NeurIPS 2017 · similarity 0.88)
- Revisiting Unbiased Implicit Variational Inference (ICML 2025 · similarity 0.86)
- Stochastic variance-reduced Gaussian variational inference on the Bures-Wasserstein manifold (ICLR 2025 · similarity 0.86)
- Particle Semi-Implicit Variational Inference (NeurIPS 2024 · similarity 0.86)
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.