Summary of Arrivalnet: Predicting City-wide Bus/tram Arrival Time with Two-dimensional Temporal Variation Modeling, by Zirui Li and Patrick Wolf and Meng Wang
ArrivalNet: Predicting City-wide Bus/Tram Arrival Time with Two-dimensional Temporal Variation Modeling
by Zirui Li, Patrick Wolf, Meng Wang
First submitted to arxiv on: 17 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- 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 paper proposes ArrivalNet, a novel two-dimensional temporal variation-based multi-step arrival time prediction (ATP) model for buses and trams. It tackles the issue of previous methods focusing solely on one-dimensional temporal information by incorporating latent periodic information within time series. This approach enables the development of more transferable and applicable prediction models for public transport management systems. The proposed architecture decomposes one-dimensional temporal sequences into intra-periodic and inter-periodic variations, which are then transformed into two-dimensional tensors (2D blocks). Each row of a tensor contains time points within a period, while each column involves time points at the same intra-periodic index across various periods. This image-like feature representation allows for effective learning with computer vision backbones. The 2D block module is designed as a basic module for flexible aggregation, drawing from the concept of residual neural networks. Additionally, contextual factors such as workdays, peak hours, and intersections are incorporated into the augmented feature representation to enhance prediction performance. Experimental results demonstrate that ArrivalNet outperforms state-of-the-art baseline methods in terms of root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Arrival time prediction for buses and trams is crucial for public transport operations. Current methods only consider one-dimensional temporal information, ignoring the important periodic patterns within time series data. This paper presents ArrivalNet, a new approach that incorporates both intra-periodic and inter-periodic variations into two-dimensional tensors. The model uses computer vision backbones to learn from these 2D blocks and also considers contextual factors like workdays and peak hours. By doing so, ArrivalNet can predict arrival times more accurately than previous methods. |
Keywords
» Artificial intelligence » Mae » Time series
