Summary of Distribution-aware Online Continual Learning For Urban Spatio-temporal Forecasting, by Chengxin Wang et al.
Distribution-aware Online Continual Learning for Urban Spatio-Temporal Forecasting
by Chengxin Wang, Gary Tan, Swagato Barman Roy, Beng Chin Ooi
First submitted to arxiv on: 24 Nov 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 A novel online continual learning framework called DOST is introduced for urban spatio-temporal (ST) forecasting, which addresses the non-stationary nature of urban ST data by employing an adaptive ST network with a variable-independent adapter. The framework also incorporates an awake-hibernate learning strategy to fine-tune the adapter during online phase and prevent catastrophic forgetting. Experimental results show that DOST outperforms state-of-the-art models on four real-world datasets, achieving a 12.89% reduction in forecast errors. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Urban ST forecasting is important for intelligent scheduling and trip planning. This paper introduces a new way to improve forecasts by learning from data as it changes over time. The approach uses an adaptive network that can adjust to new patterns and prevent forgetting old information. It’s faster and more accurate than other methods, making it useful for real-world applications. |
Keywords
» Artificial intelligence » Continual learning
