Summary of Wilsonian Renormalization Of Neural Network Gaussian Processes, by Jessica N. Howard et al.
Wilsonian Renormalization of Neural Network Gaussian Processes
by Jessica N. Howard, Ro Jefferson, Anindita Maiti, Zohar Ringel
First submitted to arxiv on: 9 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Disordered Systems and Neural Networks (cond-mat.dis-nn); High Energy Physics – Theory (hep-th); Machine Learning (stat.ML)
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 In this paper, researchers apply the renormalization group (RG) technique from theoretical physics to Gaussian Process (GP) Regression, a type of machine learning model. The RG method helps separate relevant from irrelevant information, allowing for more efficient modeling and scientific inquiry. By systematically integrating out unlearnable modes in the GP kernel, the authors derive an RG flow that sets the data as the infrared scale. This approach is analytically tractable and provides a natural connection between RG flow and learnable vs. unlearnable modes. The study may improve understanding of feature learning in deep neural networks and identify potential universality classes. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses a special tool from physics to help machine learning models work better. They take a technique called renormalization group (RG) and apply it to a type of model called Gaussian Process Regression. This helps the model focus on important information and ignore unimportant details. The authors show how this works by “integrating out” things that are hard for the model to learn, which makes the model more efficient. This research might help us understand how deep neural networks work and find patterns in their behavior. |
Keywords
» Artificial intelligence » Machine learning » Regression
