Summary of Structural Entropy Guided Probabilistic Coding, by Xiang Huang et al.
Structural Entropy Guided Probabilistic Coding
by Xiang Huang, Hao Peng, Li Sun, Hui Lin, Chunyang Liu, Jiang Cao, Philip S. Yu
First submitted to arxiv on: 12 Dec 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 The proposed probabilistic embedding, SEPC, outperforms state-of-the-art models on 12 natural language understanding tasks by incorporating structural information between latent variables through a novel entropy-guided regularization loss. This paper also introduces a probabilistic encoding tree for regression task transformation, demonstrating superior effectiveness, generalization capability, and robustness to label noise. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Probabilistic embeddings can better describe data uncertainty and complexity than deterministic ones. This paper proposes SEPC, which incorporates structural information between latent variables using an entropy-guided regularization loss. It also uses a probabilistic encoding tree for regression tasks, showing improved performance on 12 natural language understanding tasks. |
Keywords
» Artificial intelligence » Embedding » Generalization » Language understanding » Regression » Regularization
