What is it about?
Many traffic forecasting AI assumes that every road will behave the same way. We show that every road is actually unique. So we invented a new AI that take this into account. This is a modified version of the transformer algorithm, the same algorithm that serves as the backbone of the popular chatGPT.
Featured Image
Photo by Robin Pierre on Unsplash
Why is it important?
Underneath chatGPT is a word forecasting model, it simply forecast the next word. Yet it has many different important applications. Similarly, improving traffic forecasting models would have many important downstream applications as well.
Perspectives
This article exemplify the kind of work I love doing. That is getting insights from the data, and then designing the appropriate deep learning methods accordingly.
Arian Prabowo
RMIT University
Read the Original
This page is a summary of: Because Every Sensor Is Unique, so Is Every Pair: Handling Dynamicity in Traffic Forecasting, May 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3576842.3582362.
You can read the full text:
Contributors
The following have contributed to this page
