Winners of the Traffic4cast 2020 Core Competition

The Institute of Advanced Research in Artificial Intelligence (IARAI) is proud to announce the winners of the Traffic4cast 2020 core competition. The goal of the competition is to predict traffic in multiple big cities of different culture and economy based on industrial-scale real-world traffic data. Traffic4cast is chosen again this year for the prestigious Neural Information Processing Systems (NeurIPS), the leading event in machine learning. Three top-ranked teams in the core competition leaderboard, selected by rigorous statistical analysis, will receive highly competitive awards.

The winners are:

1^st prize: Sungbin Choi. Award: 10 000 USD value and 12 month Research Fellowship at IARAI, covering stipend and expenses.

2^nd prize: TLabV2 team represented by Fanyou Wu¹, Yang Liu², Zhiyuan Liu², Xiaobo Qu³, Rado Gazo¹, and Eva Haviarova¹ from ¹Purdue University, USA, ²Southeast University, China, and ³Chalmers University of Technology, Sweden. Award: 5 000 USD value and 12 month Research Fellowship at IARAI, covering stipend and expenses.

3^rd prize: LDS Group represented by Jingwei Xu¹, Jianjin Zhang², Zhiyu Yao³, and Yunbo Wang¹ from ¹Shanghai Jiao Tong University, China, ²Microsoft, Beijing, China, and ³Tsinghua University, China. Award: 2 000 USD value and complimentary registration for NeurIPS 2021.

Congratulations!

Each of the winners will give a highlight talk at the NeurIPS 2020 conference. Traffic4cast Scientific Committee has selected two additional highlight talks from the submitted abstracts based on their exceptional scientific input.

The selected presentations for the Traffic4cast session at NeurIPS 2020 competition track are:

1. “Utilizing UNet for the future traffic map prediction“, Sungbin Choi.

2. “TLab: Traffic Map Movie Forecasting Based on HR-NET“, Fanyou Wu, Purdue University, USA.

3. “Towards Good Practices of U-Net for Traffic Forecasting“, Jingwei Xu, Shanghai Jiao Tong University, China.

4. “Graph Ensemble Net and the Importance of Feature & Loss Function Design for Traffic Prediction”, Qi Qi.

5. “Uncertainty Intervals for Graph-based Spatio-Temporal Traffic Prediction”, Tijs Maas, University of Amsterdam, The Netherlands.