Summary of Gder: Safeguarding Efficiency, Balancing, and Robustness Via Prototypical Graph Pruning, by Guibin Zhang et al.
GDeR: Safeguarding Efficiency, Balancing, and Robustness via Prototypical Graph Pruning
by Guibin Zhang, Haonan Dong, Yuchen Zhang, Zhixun Li, Dingshuo Chen, Kai Wang, Tianlong Chen, Yuxuan Liang, Dawei Cheng, Kun Wang
First submitted to arxiv on: 17 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 novel dynamic soft-pruning method, GDeR, is introduced to address the issue of data pruning for graph neural networks (GNNs). GDeR constructs a well-modeled graph embedding hypersphere and samples representative, balanced, and unbiased subsets from this space. The approach achieves or surpasses the performance of the full dataset with 30%~50% fewer training samples, attains up to a 2.81x lossless training speedup, and outperforms state-of-the-art pruning methods in imbalanced training and noisy training scenarios. The method is designed to maintain the efficiency of previous data pruning practices while ensuring balance and robustness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary GNNs are special types of artificial intelligence models that work with graph structures. Right now, these models are having trouble working with really big datasets because they’re not good at dealing with imbalanced or noisy data. This is a problem because it makes it hard to train the models quickly and accurately. To fix this, scientists have developed a new way to prune data called GDeR. It’s like a special filter that helps get rid of unnecessary data while keeping the important parts. This method has been tested on five different datasets and showed some really good results. It was able to train the models faster and more accurately than previous methods. |
Keywords
* Artificial intelligence * Embedding * Pruning
