Summary of Dirichlet-based Coarse-to-fine Example Selection For Open-set Annotation, by Ye-wen Wang et al.
Dirichlet-Based Coarse-to-Fine Example Selection For Open-Set Annotation
by Ye-Wen Wang, Chen-Chen Zong, Ming-Kun Xie, Sheng-Jun Huang
First submitted to arxiv on: 26 Sep 2024
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 This paper proposes a novel approach to active learning, dubbed Dirichlet-based Coarse-to-Fine Example Selection (DCFS), which addresses the limitations of traditional active learning methods in real-world scenarios. The proposed strategy, DCFS, leverages simplex-based evidential deep learning (EDL) to break translation invariance and distinguish between known and unknown classes by incorporating evidence-based data and distribution uncertainty. Additionally, hard known-class examples are identified using model discrepancy generated from two classifier heads, where the authors amplify and alleviate this discrepancy for unknown and known classes, respectively. The paper combines these components to form a two-stage strategy that selects the most informative examples from known classes. Experimental results on various openness ratio datasets demonstrate that DCFS achieves state-of-the-art performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine you’re trying to learn something new by asking questions and getting answers. But sometimes, the answers are not correct or make no sense. This paper tries to solve this problem in a special kind of learning called active learning. They propose a new way to pick the most important information from what we already know, so that we can learn better. It’s like using a special filter to get rid of noise and distractions. The authors tested their method on many different datasets and showed that it works really well. |
Keywords
» Artificial intelligence » Active learning » Deep learning » Translation
