Summary of Provably Better Explanations with Optimized Aggregation Of Feature Attributions, by Thomas Decker et al.
Provably Better Explanations with Optimized Aggregation of Feature Attributions
by Thomas Decker, Ananta R. Bhattarai, Jindong Gu, Volker Tresp, Florian Buettner
First submitted to arxiv on: 7 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); 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 This paper explores ways to improve the reliability of post-hoc explanations for machine learning models by combining multiple feature attribution methods. Despite numerous techniques available, current approaches often produce inconsistent results, raising concerns about their overall quality. The authors propose a novel approach to derive optimal combinations of feature attributions that meet desired criteria such as robustness or faithfulness to model behavior. Through experiments with various architectures and popular methods, the combination strategy outperforms individual methods and existing baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Machine learning models are becoming increasingly powerful, but it can be hard to understand how they make predictions. This paper looks at ways to explain these predictions after they’ve been made. Right now, there are many different techniques for doing this, but each one has its own problems. The authors of this paper want to find a way to combine these techniques so that the explanations are more reliable and accurate. They propose a new method for doing this and test it on different models and datasets. The results show that their approach is better than just using one technique alone. |
Keywords
» Artificial intelligence » Machine learning
