Summary of Open Knowledge Base Canonicalization with Multi-task Unlearning, by Bingchen Liu et al.
Open Knowledge Base Canonicalization with Multi-task Unlearning
by Bingchen Liu, Shihao Hou, Weixin Zeng, Xiang Zhao, Shijun Liu, Li Pan
First submitted to arxiv on: 25 Oct 2023
Categories
- Main: Artificial Intelligence (cs.AI)
- 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 proposed paper addresses the issue of constructing large open knowledge bases (OKBs) in mobile computing, where noun phrases and relational phrases often suffer from redundancy and ambiguity. To tackle this problem, the authors investigate OKB canonicalization, which requires removing sensitive information or outdated data to meet privacy regulations and ensure timely data. They propose a multi-task unlearning framework, MulCanon, that utilizes noise characteristics in the diffusion model for machine unlearning in OKB canonicalization. The framework unifies learning objectives of diffusion models, knowledge graph embedding (KGE), and clustering algorithms using a two-step multi-task learning paradigm. Experimental results on popular OKB canonicalization datasets show that MulCanon achieves advanced machine unlearning effects. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary OKBs are like huge libraries of information that can be accessed from mobile devices. The problem is that this information often gets repeated or is hard to understand. To solve this, the researchers created a way to “clean up” the data by removing sensitive information and outdated facts. They call this process OKB canonicalization. To do this, they use machine learning techniques like clustering and knowledge graph embedding (KGE). The goal is to make sure that the cleaned-up data is accurate and up-to-date. |
Keywords
* Artificial intelligence * Clustering * Diffusion model * Embedding * Knowledge graph * Machine learning * Multi task
