Summary of Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction, by Hamed Karimi and Reza Samavi
Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction
by Hamed Karimi, Reza Samavi
First submitted to arxiv on: 16 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); 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 The proposed Evidential Conformal Prediction (ECP) method generates conformal prediction sets for image classifiers by leveraging a non-conformity score function rooted in Evidential Deep Learning (EDL). This approach quantifies model uncertainty in DNN classifiers using evidence derived from logit values of target labels. The non-conformity score is comprised of three components: heuristic notion of uncertainty, uncertainty surprisal, and expected utility. Experimental results show that ECP outperforms three state-of-the-art methods for generating CP sets in terms of set sizes, adaptivity, and coverage of true labels. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Evidential Conformal Prediction is a new way to predict how confident an image classifier is about its answers. This helps make sure the predictions are correct and reliable. The method uses special math to figure out how uncertain the model is based on what it knows. It also compares this uncertainty to what’s actually happening in the images. In tests, this approach did better than other ways of doing conformal prediction. |
Keywords
* Artificial intelligence * Deep learning
