Summary of Time Series Representation Learning with Supervised Contrastive Temporal Transformer, by Yuansan Liu et al.
Time Series Representation Learning with Supervised Contrastive Temporal Transformer
by Yuansan Liu, Sudanthi Wijewickrema, Christofer Bester, Stephen O’Leary, James Bailey
First submitted to arxiv on: 16 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 research proposes a novel fusion model called SCOTT (Supervised Contrastive Temporal Transformer) for learning effective representations of time series data. The approach combines techniques from representation learning and time series analysis, including augmentation methods, Transformers, and Temporal Convolutional Networks. The authors evaluate SCOTT on the task of Time Series Classification using 45 datasets from the UCR archive, achieving state-of-the-art performance or comparable results to existing models. They also apply SCOTT to a real-world problem, online Change Point Detection (CPD), demonstrating high reliability and efficiency. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary SCOTT is a new way to learn useful representations of time series data. The model uses a combination of techniques to help it understand patterns in the data. It’s tested on many different datasets and does well compared to other models. SCOTT also works well for a specific task called Change Point Detection, which helps identify when something important happens in a sequence of data. |
Keywords
* Artificial intelligence * Classification * Representation learning * Supervised * Time series * Transformer
