Yang Song | Sliced Score Matching: A Scalable Approach to Density and Score Estimation
yang-song.github.io · 1,274 words · saved by 1 readers
Yang Song's academic website.
Sliced Score Matching: A Scalable Approach to Density and Score Estimation | Yang Song Sliced Score Matching: A Scalable Approach to Density and Score Estimation An overview for our UAI 2019 paper on Sliced Score Matching. We show how to use random projections to scale up score matching—a classic method to learn unnormalized probabilisic models—to high-dimensional data. Theoretically, sliced score matching produces a consistent and asymptotic normal estimator under some regularity conditions. We apply sliced score matching to training deep energy-based models, learning VAEs with implicit encod
saved by
related reading
- Generative Modeling by Estimating Gradients of the Data Distribution | Yang Songyang-song.github.io
- Diffusion Models as a kind of VAE | Angus Turnerangusturner.github.io
- Fabian Fuchsfabianfuchsml.github.io
- ML vs. Score matchingarxiv.org
- Regularized by Score Matching (LeCun)papers.nips.cc
- Generative Modeling by Estimating Gradients of the Data Distribution | Yang Songyang-song.github.io
- Deep Energy Estimator.pdfarxiv.org
- score matching (hyvarinen)jmlr.org
- What are Diffusion Models? | Lil'Loglilianweng.github.io
- Denoising Density Estimatorsarxiv.org
- Yang Songyang-song.net
- Riemannian Metric Matching for Scalable Geometric Modeling of Distributionsarxiv.org