Summary of Optimizing Sequential Recommendation Models with Scaling Laws and Approximate Entropy, by Tingjia Shen et al.
Optimizing Sequential Recommendation Models with Scaling Laws and Approximate Entropy
by Tingjia Shen, Hao Wang, Chuhan Wu, Jin Yao Chin, Wei Guo, Yong Liu, Huifeng Guo, Defu Lian, Ruiming Tang, Enhong Chen
First submitted to arxiv on: 30 Nov 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Information Retrieval (cs.IR)
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 Performance Law for Sequential Recommendation (SR) models addresses the challenges of scalability by investigating the relationship between model performance and data quality. The authors introduce Approximate Entropy (ApEn) as a novel metric to assess data quality, which is more nuanced than traditional quantity metrics. By fitting HR and NDCG metrics to transformer-based SR models, the study demonstrates a strong correlation in large SR models, offering insights into achieving optimal performance for any given model configuration. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Sequential Recommendation (SR) models need to be optimized to handle increasing data sizes. Researchers have discovered Scaling Laws that explain how model performance changes with size. But these laws don’t work well with recommender systems because they have structural and collaborative problems. The authors of this paper introduce a new Performance Law for SR models, which helps understand the relationship between model performance and data quality. They use HR and NDCG metrics to fit their model and show that it works well on large datasets. |
Keywords
» Artificial intelligence » Scaling laws » Transformer
