Summary of Global Versus Local: Evaluating Alexnet Architectures For Tropical Cyclone Intensity Estimation, by Vikas Dwivedi
Global versus Local: Evaluating AlexNet Architectures for Tropical Cyclone Intensity Estimation
by Vikas Dwivedi
First submitted to arxiv on: 11 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); Atmospheric and Oceanic Physics (physics.ao-ph)
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 two ensemble-based models based on AlexNet architecture for estimating tropical cyclone intensity using visible satellite images. The global model is trained on the entire dataset, while the distributed model trains multiple AlexNets on subsets of the training data categorized by Saffir-Simpson wind speed scale. Both models outperform a benchmark model called Deepti, with RMSEs of 9.03 and 9.3 knots respectively. The paper also discusses the performance of AlexNet using gradient class activation maps (grad-CAM). This work allows for future experimentation with various deep learning models in single and multi-channel settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers developed two new ways to detect the intensity of tropical cyclones from satellite images. They used a type of artificial intelligence called AlexNet, which is already good at recognizing objects like cats and dogs. The researchers trained two different versions of AlexNet: one that learned from all the data together, and another that learned from smaller pieces of data grouped by how strong the winds were. Both versions did better than a previous model called Deepti. This new approach allows scientists to test other types of artificial intelligence models on this problem. |
Keywords
» Artificial intelligence » Deep learning
