Summary of Class-balanced and Reinforced Active Learning on Graphs, by Chengcheng Yu et al.
Class-Balanced and Reinforced Active Learning on Graphs
by Chengcheng Yu, Jiapeng Zhu, Xiang Li
First submitted to arxiv on: 15 Feb 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 In this research paper, the authors propose a novel active learning framework for graph neural networks (GNNs) called GCBR (Graph Class-Balanced Reinforced). The goal is to query valuable samples from unlabeled data for annotation, maximizing GNN performance at a lower cost. To address the issue of class imbalance in GNN training, GCBR learns an optimal policy to acquire class-balanced and informative nodes for annotation. The approach uses reinforcement learning with an Advantage Actor-Critic (A2C) algorithm and incorporates a punishment mechanism to obtain a more class-balanced labeled set. Experimental results on multiple datasets demonstrate the effectiveness of GCBR and its upgraded version, GCBR++, which achieves superior performance over state-of-the-art baselines. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary GNNs are special kinds of artificial intelligence that can learn from graphs. Graphs are like maps that show how things are connected to each other. The goal is to make GNNs better by choosing the right data points to label, so they can learn more efficiently. Right now, there’s a problem with this process because it can lead to biased results if some classes have way more examples than others. To fix this, researchers created a new approach called GCBR that helps GNNs learn from a balanced mix of examples. This means the AI is more likely to make good decisions. The team tested their idea on several datasets and found that it works better than other approaches. |
Keywords
* Artificial intelligence * Active learning * Gnn * Reinforcement learning
