Summary of Producing Plankton Classifiers That Are Robust to Dataset Shift, by Cheng Chen et al.
Producing Plankton Classifiers that are Robust to Dataset Shift
by Cheng Chen, Sreenath Kyathanahally, Marta Reyes, Stefanie Merkli, Ewa Merz, Emanuele Francazi, Marvin Hoege, Francesco Pomati, Marco Baity-Jesi
First submitted to arxiv on: 25 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
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 This study focuses on developing more robust deep learning classifiers for species recognition in water ecosystems, a crucial task for modern plankton high-throughput monitoring. The researchers investigate the issue of Dataset Shift, which causes performance drops during deployment. They propose a three-step pipeline to identify potential pitfalls when classifying new data and pinpoint features that impact classification. The study also presents a robust model called BEsT, an ensemble of BEiT vision transformers with targeted augmentations, achieving an 83% OOD accuracy. This work contributes to the development of reliable plankton classification technologies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us understand how to make machines better at recognizing different types of plankton in water. Right now, these machines are good at identifying plankton when they’re looking at normal pictures, but they get confused when shown new pictures that are a bit different. The researchers looked into why this happens and found some ways to improve the machines so they can work better even with new pictures. They also came up with a special kind of machine called BEsT that does really well at recognizing plankton. |
Keywords
* Artificial intelligence * Classification * Deep learning
