|$100K Machine Learning Challenge|
|Written by Lucy Black|
|Tuesday, 24 May 2016|
In partnership with Udacity, Chinese ridesharing company Didi Chuxing has announced a global machine learning competition to come up with a better ride matching algorithm using Didi’s published data sets.
The grand prize winner not only gets $100,000 and worldwide acclaim, and also a chance to work at the Didi Research Lab in China or Silicon Valley. The judging panel includes Sebastian Thrun, founder of Udacity and Peter Norvig, Director of Research at Google as well as He Xiaofei, President of the DIDI Research Institute.
Less than 10% of China's 1.4 billion citizens own automobiles so there is a big demand for ridesharing services. Didi Chuxing processes over 11 million trips, plans over 9 billion routes and collects over 50TB of data per day. The rationale for this contest, which is to improve supply-demand forecasting, is therefore obvious:
To meet needs of riders, Didi must continually innovate to improve cloud computing and big data technologies and algorithms in order to process this massive amount of data and uphold service reliability.
The Udacity blog provides this explanation:
Machine learning strategies are vital to the company’s success, and with growth comes the need to constantly improve on core algorithms, especially those that impact supply-demand forecasting. The competition is a challenge to machine learning and big data students around the world to improve how the company ensures riders always get a car when and where they need it, and drivers know where to be even before a ride is hailed.
The contest, which offers three large cash prizes, is now open and the competition data set is available together with the evaluation metrics. The deadline to register and submit ride-matching algorithms for the first round is June 17th at 8:30 PM PT. Teams can make as many submissions as they wants and the results of the first round will be on historical best submissions.
While there is a public leaderboard, updated on a daily basis, it will be switched off a week before the deadline and will not count towards the final ranking. Meanwhile being on the leaderboard does count - and over 250 teams already are - the leading team in the first week (on May 26th at 12:00 PM China Standard time) will win a new MacBook.
The top 50 teams from the first round will proceed to the final round which runs from June 21 to July 11. The top 10 teams will be invited to attend an on-site event in Bejing (expenses paid) to defend their solutions. As well as the three cash prizes other awards such as Best Algorithm and Best Presentation will be awarded to the teams participating in the on-site stage.
or email your comment to: firstname.lastname@example.org
|Last Updated ( Tuesday, 24 May 2016 )|