Summary of Can Biases in Imagenet Models Explain Generalization?, by Paul Gavrikov and Janis Keuper
Can Biases in ImageNet Models Explain Generalization?
by Paul Gavrikov, Janis Keuper
First submitted to arxiv on: 1 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); 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 research paper, the authors investigate how deep learning models generalize to rare in-distribution and out-of-training-distribution samples. They explore biases that differentiate models from human vision and their impact on generalization. The study uses ResNet-50 architecture and 48 ImageNet models obtained via different training methods to analyze the interaction between these biases and generalization. The findings suggest that the previously identified biases are insufficient to accurately predict a model’s generalization. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers aim to understand why deep learning models struggle with generalizing to rare samples. They focus on biases that make models behave differently than human vision and investigate how these biases affect performance. By analyzing 48 ImageNet models, they find that previously identified biases don’t accurately predict a model’s ability to generalize. |
Keywords
» Artificial intelligence » Deep learning » Generalization » Resnet
