Summary of Dynamic Evidence Decoupling For Trusted Multi-view Learning, by Ying Liu et al.
Dynamic Evidence Decoupling for Trusted Multi-view Learning
by Ying Liu, Lihong Liu, Cai Xu, Xiangyu Song, Ziyu Guan, Wei Zhao
First submitted to arxiv on: 4 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 trusted multi-view learning method, CCML, aims to improve decision accuracy and uncertainty estimation by learning class distributions for each instance. The method constructs view opinions using evidential deep neural networks, which consist of belief mass vectors and uncertainty estimates. The consistent evidence is derived from shared portions across all views, while the complementary evidence is obtained by averaging differing portions. The opinion constructed from consistent evidence strictly aligns with ground-truth categories, allowing for potential vagueness in complementary evidence. CCML outperforms state-of-the-art baselines on accuracy and reliability on one synthetic and six real-world datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary CCML is a new way to learn from multiple sources of information that helps make better decisions by considering how sure we are about each decision. This is important because some data might not be clear or reliable, which can affect the results. The method uses special kinds of neural networks to combine different views and opinions. It also separates consistent and complementary evidence, allowing for more accurate and uncertain decisions. |
