Summary of Beyond the Black Box: Do More Complex Deep Learning Models Provide Superior Xai Explanations?, by Mateusz Cedro et al.
Beyond the Black Box: Do More Complex Deep Learning Models Provide Superior XAI Explanations?
by Mateusz Cedro, Marcin Chlebus
First submitted to arxiv on: 14 May 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
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 study investigates the relationship between deep learning model complexity, Explainable AI (XAI) efficacy, and classification performance in lung X-ray images of COVID-19-infected and healthy patients. Four ResNet architectures (ResNet-18, 34, 50, 101) are utilized to evaluate models’ classification accuracy, AUC-ROC scores, and the relevance of corresponding XAI explanations. The results show that increased model complexity is associated with decreased classification accuracy and AUC-ROC scores, but surprisingly, no significant differences occur in XAI quantitative metrics between trained models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The research uses deep learning models and Explainable AI to analyze lung X-ray images and diagnose COVID-19. They test four different models (ResNet-18, 34, 50, 101) to see how well they work and if the explanations for their decisions are good or bad. The study finds that making the models more complex doesn’t always make them better at diagnosing COVID-19 or making good explanations. |
Keywords
» Artificial intelligence » Auc » Classification » Deep learning » Resnet
