Summary of Disensemi: Semi-supervised Graph Classification Via Disentangled Representation Learning, by Yifan Wang et al.
DisenSemi: Semi-supervised Graph Classification via Disentangled Representation Learning
by Yifan Wang, Xiao Luo, Chong Chen, Xian-Sheng Hua, Ming Zhang, Wei Ju
First submitted to arxiv on: 19 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Social and Information Networks (cs.SI)
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 proposed framework, DisenSemi, tackles semi-supervised graph classification by learning disentangled representations for both supervised and unsupervised models. The approach generates factor-wise graph representations using a disentangled graph encoder and trains two models via supervised objectives and mutual information (MI)-based constraints. To ensure meaningful knowledge transfer from the unsupervised encoder to the supervised one, MI-based disentangled consistency regularization is defined. Experimental results on various datasets demonstrate the effectiveness of DisenSemi. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers created a new way to help machines learn about different types of data, like pictures and videos. They wanted to make it work even when there isn’t enough labeled data available. This is called semi-supervised learning. The idea is to use both labeled and unlabeled data to train the machine. The scientists came up with a new method called DisenSemi that separates the information into different parts. This helps the machine learn more accurately. They tested it on some big datasets and showed that it works well. |
Keywords
» Artificial intelligence » Classification » Encoder » Regularization » Semi supervised » Supervised » Unsupervised
