Summary of Faster Sampling Via Stochastic Gradient Proximal Sampler, by Xunpeng Huang et al.
Faster Sampling via Stochastic Gradient Proximal Sampler
by Xunpeng Huang, Difan Zou, Yi-An Ma, Hanze Dong, Tong Zhang
First submitted to arxiv on: 27 May 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper investigates Stochastic Proximal Samplers (SPS) for sampling from non-log-concave distributions. The authors develop a general framework for implementing SPS and establish convergence theory, showing that the algorithm can guarantee convergence to the target distribution as long as the second moment of the algorithm trajectory is bounded and restricted Gaussian oracles can be well approximated. Two implementable variants are proposed based on Stochastic gradient Langevin dynamics (SGLD) and Metropolis-adjusted Langevin algorithm (MALA), which outperform previous results by at least an O(d^{1/3}) factor. The authors demonstrate the efficiency of their proposed algorithm through empirical studies on synthetic data with various dimensions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at a new way to sample from big datasets, called Stochastic Proximal Samplers (SPS). It shows how SPS can work efficiently and accurately for sampling from certain types of distributions. The researchers propose two different ways to implement SPS and show that it’s better than previous methods by a lot. They test their idea on fake data and it works well. |
Keywords
» Artificial intelligence » Synthetic data
