Summary of A Rationale From Frequency Perspective For Grokking in Training Neural Network, by Zhangchen Zhou et al.
A rationale from frequency perspective for grokking in training neural network
by Zhangchen Zhou, Yaoyu Zhang, Zhi-Qin John Xu
First submitted to arxiv on: 24 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Neural and Evolutionary Computing (cs.NE); 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 The paper provides an empirical frequency perspective to explain the emergence of the “grokking” phenomenon in neural networks (NNs), where NNs initially fit the training data and later generalize to the test data during training. The authors observe this phenomenon across synthetic and real datasets, offering a novel viewpoint for elucidating the grokking phenomenon by characterizing it through the lens of frequency dynamics during the training process. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This phenomenon is also known as “grokking,” where neural networks initially fit the training data and later generalize to the test data. The researchers found that neural networks learn the less salient frequency components present in the test data, which helps them generalize better. They tested this on both synthetic and real datasets. |
