Summary of Learning to Compress: Local Rank and Information Compression in Deep Neural Networks, by Niket Patel et al.
Learning to Compress: Local Rank and Information Compression in Deep Neural Networks
by Niket Patel, Ravid Shwartz-Ziv
First submitted to arxiv on: 10 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Information Theory (cs.IT)
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 Deep neural networks exhibit a bias towards low-rank solutions during training, implicitly learning low-dimensional feature representations. This paper investigates how deep multilayer perceptrons (MLPs) encode these feature manifolds, connecting this behavior to the Information Bottleneck (IB) theory. The authors introduce local rank as a measure of feature manifold dimensionality and demonstrate theoretically and empirically that this rank decreases during the final phase of training. They argue that networks reducing the rank of their learned representations also compress mutual information between inputs and intermediate layers. This work bridges the gap between feature manifold rank and information compression, offering new insights into the interplay between information bottlenecks and representation learning. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Deep neural networks have a secret: they tend to simplify features during training. This paper looks at how deep neural networks do this and why it’s important. The authors find that as these networks learn, they reduce the number of features they use. They also show that when networks do this, they compress information between what we put in and what they produce. This is interesting because it helps us understand how these networks really work. |
Keywords
» Artificial intelligence » Representation learning
