Summary of Imagenot: a Contrast with Imagenet Preserves Model Rankings, by Olawale Salaudeen and Moritz Hardt
ImageNot: A contrast with ImageNet preserves model rankings
by Olawale Salaudeen, Moritz Hardt
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
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 introduces ImageNot, a new dataset designed to match the scale of ImageNet while differing significantly from it. The study shows that key model architectures developed for ImageNet over the years perform similarly when trained and evaluated on ImageNot as they do on ImageNet, both from scratch and fine-tuning models. Moreover, the relative improvements of each model over earlier models strongly correlate in both datasets. Additionally, the paper demonstrates the utility of ImageNot for transfer learning purposes, highlighting a surprising degree of external validity in the relative performance of image classification models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The study shows that some computer models are surprisingly good at classifying images, even when trained on different pictures. This is important because it means that these models can be used to help identify objects in new and different photos. The researchers created a new dataset called ImageNot, which has many similarities to the popular ImageNet dataset. They found that models developed for ImageNet work just as well on ImageNot, whether they’re trained from scratch or built upon earlier models. This is great news because it means that these models can be used in a variety of situations. |
Keywords
* Artificial intelligence * Fine tuning * Image classification * Transfer learning
