Summary of Coke: Causal Discovery with Chronological Order and Expert Knowledge in High Proportion Of Missing Manufacturing Data, by Ting-yun Ou et al.
COKE: Causal Discovery with Chronological Order and Expert Knowledge in High Proportion of Missing Manufacturing Data
by Ting-Yun Ou, Ching Chang, Wen-Chih Peng
First submitted to arxiv on: 17 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Methodology (stat.ME)
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 proposes a novel approach, COKE, to construct causal graphs in manufacturing datasets by leveraging expert knowledge and chronological order among sensors. The method maximizes the use of samples with missing values, derives embeddings from intersections with an initial graph that incorporates expert knowledge and chronological order, and creates a sensor ordering graph. The graph-generating process is optimized using an actor-critic architecture to obtain a final graph with a maximum reward. Experimental evaluations demonstrate that COKE outperforms benchmark methods by 39.9% in the F1-score on average, with improvements reaching 62.6% for real-world semiconductor datasets and 85.0% in configurations similar to real-world datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, researchers developed a new way to analyze data from machines in factories. They wanted to find relationships between what’s happening at different sensors. The problem is that the data often has missing values, which makes it hard to figure out what’s going on. The researchers also had to use special knowledge from experts and take into account when things happened in order. They created a new method called COKE that uses all this information to make better predictions. This approach worked really well, improving results by 39.9% on average. |
Keywords
» Artificial intelligence » F1 score
