Banner Before Header

Traffic4cast Competition Calls on AI Community to Better Predict Urban Traffic Flows

1 316

Second annual competition for machine learning experts to improve planning tools for urban mobility

Top ranked teams receive prize money, research fellowships, honors at NeurIPS 2020

Vienna – The Institute for Advanced Research in Artificial Intelligence (IARAI), an independent and global machine-learning research institute established by HERE Technologies, announced the second annual Traffic4Cast competition. This year’s competition for the AI community expands the datasets and number of cities being analyzed. Participants must create algorithms that predict future traffic flows for at least 10 cities.

Marketing Technology News: Adapty Delivers First Ever Insite Online B2B Commerce Implementation

The success of Traffic4cast 2019 resulted in more than 40 teams competing around the world with winners from South Korea, Oxford/Zurich and Toronto. All top-ranked teams demonstrated that neural networks were the most effective method used at predicting traffic instead of “non-black box” solutions. The winning Traffic4cast 2019 solutions are being published in a special NeurIPS volume of the ‘Proceedings of Machine Learning Research’ journal.

Submissions for this year’s competition are due by October 15, 2020. The top-ranked teams will be honored at the NeurIPS 2020 conference, will have an opportunity to see their solutions published in a special NeurIPS volume of the ‘Proceedings of Machine Learning Research’ journal, and receive:

  • 1st prize: $10,000 USD value and 12-month Research Fellowship at IARAI, covering stipend and expenses;
  • 2nd prize: $5,000 USD value and 12-month Research Fellowship at IARAI, covering stipend and expenses;
  • 3rd prize: $2,000 USD value and complimentary registration for NeurIPS 2020
  • 1st – 6th prizes: complimentary registration for NeurIPS 2020.

In collaboration with HERE, IARAI is providing participants with an unprecedented amount of traffic data for at least 10 entire cities through the course of a full year. The data has been transformed into high-definition movie clips that, frame by frame, depict traffic over time, including morning, evening and rush hour traffic events.

New for this year’s competition, the traffic movie clips will be augmented by static and dynamic features, such as points of interest, weather, air pollution, and special events. The bonus challenge invites participants to explore the effects these additional features have on predictive traffic models.

Marketing Technology News:  Druva and FireEye Join Forces to Strengthen Enterprise Cyber Resiliency

“We’re excited to call upon the global AI community to take on the Traffic4Cast 2020 competition,” said Sepp Hochreiter, a founding co-director of IARAI and an artificial intelligence pioneer, who invented the long short-term memory (LSTM) neural network architecture. “Last year, teams demonstrated novel approaches to understanding complex traffic systems based on applications of neural networks. Building on this excitement and a real opportunity to improve urban mobility, this year’s competition is focused on going deeper and to the next level in understanding complex traffic systems.”

“Following the success of last year’s IARAI Traffic4Cast competition results, it is clear that neural network architectures will play an increasingly important role in terms of predicting future traffic patterns and making mobility smarter and cleaner,” said Giovanni Lanfranchi, Senior Vice President Development and Chief Technology Officer at HERE Technologies. “By leveraging extensive HERE traffic datasets and location data, our hope is that this year’s competition participants will be even better equipped with the tools they need to attain a higher level of traffic prediction accuracy in their models.”

Marketing Technology News: Claro Enterprise Solutions and The CMO Club Partner to Produce “Social Media Playbook”

1 Comment
  1. […] post Traffic4cast Competition Calls on AI Community to Better Predict Urban Traffic Flows appeared first on MarTech […]

Leave A Reply

Your email address will not be published.