Summary of Path Choice Matters For Clear Attribution in Path Methods, by Borui Zhang et al.
Path Choice Matters for Clear Attribution in Path Methods
by Borui Zhang, Wenzhao Zheng, Jie Zhou, Jiwen Lu
First submitted to arxiv on: 19 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 paper addresses the issue of ambiguity in Deep Neural Network (DNN) interpretations, which hampers human trust. The authors introduce the Concentration Principle to allocate high attributions to indispensable features, achieving aesthetic and sparsity. They also propose SAMP, a model-agnostic interpreter that efficiently searches near-optimal paths from pre-defined manipulation paths. To further improve rigorousness and optimality, the authors suggest infinitesimal constraint (IC) and momentum strategy (MS). The paper demonstrates the effectiveness of SAMP through visualizations that pinpoint salient image pixels, outperforming counterparts in quantitative experiments. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us understand how to make sure deep learning models are working correctly. Right now, it’s hard to see what features of an image or data point are most important for a model to make predictions. The authors come up with a new way to do this, called the Concentration Principle, which makes the results more clear and easier to understand. They also create a tool called SAMP that can be used with any deep learning model to figure out what’s most important. This helps us trust models more by making them more transparent. |
Keywords
* Artificial intelligence * Deep learning * Neural network
