Summary of Generating Robust Counterfactual Witnesses For Graph Neural Networks, by Dazhuo Qiu et al.
Generating Robust Counterfactual Witnesses for Graph Neural Networks
by Dazhuo Qiu, Mengying Wang, Arijit Khan, Yinghui Wu
First submitted to arxiv on: 30 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Databases (cs.DB)
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 introduces Robust Counterfactual Witnesses (RCWs) for Graph Neural Networks (GNNs). RCWs provide robust explanations for GNN-based node classification by identifying the fraction of a graph that remains counterfactual and factual after perturbing up to k node pairs. The authors establish hardness results, from tractable to co-NP-hardness, for verifying and generating RCWs. They also present efficient algorithms for verification and generation, as well as a parallel algorithm for large graphs with scalability guarantees. Experimental results showcase the effectiveness of RCWs on benchmark datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us understand why Graph Neural Networks make certain predictions. It introduces a new way to explain these predictions called Robust Counterfactual Witnesses. These explanations show which parts of the graph are important and why they lead to certain conclusions. The authors also show that their method is efficient and can be used for big graphs. They test it on some famous datasets and show that it works well. |
Keywords
» Artificial intelligence » Classification » Gnn
