1st Call for Participation - KDD CUP 2019
Hugo Jair
[Apologies for multiple postings]
KDD Cup 2019 - AutoML for Temporal Relational Data Track
Provided and Sponsored by 4Paradigm, ChaLearn and Microsoft
https://www.4paradigm.com/competition/kddcup2019
The KDD Cup is one of the most prestigious and long-term running competition programs, the first KDD Cup being organized more than 20 years ago. For the KDD Cup 2019 we are organizing a competition on Automatic Machine Learning (AutoML) for Temporal Relational Data. In this challenge we challenge KDD and ML researchers and practitioners to develop AutoML solutions to binary classification problems based on temporal relational data.
Temporal relational data is ubiquitous in industrial machine learning applications, such as online advertising, recommender systems, financial market analysis, etc. In this sort of data there are timestamps to indicate the timings of events and multiple related tables to provide different perspectives, such data contains useful information that can be exploited to improve machine learning performance. However, currently, the exploitation of temporal relational data is often carried out by experienced human experts with in-depth domain knowledge in a labor-intensive trial-and-error manner. The organized challenge aims at boosting research on AutoML methods for temporal relational data.
*** Participation ***
Participants of this challenge will submit code submissions for solving classification problems starting from temporal relational data. The provided datasets are arranged in multiple tables and obeying a temporal order. During the first phase of the competition, code will be run autonomously on five public datasets, and performance will be displayed in a leaderboard. Test labels will not be released to participants. In the final phase, the last code submission from each participant will be automatically evaluated in fresh new private datasets without any human intervention. Nothing of private datasets would be released. Participants will have access to the same computer resources, the top ranked participants will be eligible to prizes.
The challenge is hosted in the CodaLab platform.To participate, please visit the competition website at:
https://competitions.codalab.org/competitions/21948
and follow the instructions to learn details about the problem setup, data, submission interface, and evaluation protocol. We provide participants with a starting-kit which includes the demo data, a baseline method, and all resources needed for participants to simulate the running environment locally on their own computers.
*** Important dates *** (All times are in UTC time)
* April 1st, 2019: Beginning of the competition, release of public datasets. Participants can start submitting code and obtaining immediate feedback in the leaderboard.
* June 27th, 2019: End of the Feedback Phase, code from feedback phase is automatically migrated to the Check Phase. Check phase starts.
* July 7th, 2019: End of the Check Phase, organizers start code verification.
* July 11th, 2019: Deadline for submitting fact sheets.
* July 16th, 2019: End of the challenge, beginning of the post-competition process
* Jul 20th, 2019: Announcement of the KDD Cup Winners. (Tentative, we might choose to release the winners at KDD 2019)
* Aug 4th, 2019: KDD 2019.
*** Prizes ***
The following prizes will be offered to the top ranked participants.
1st Place: $15,000
2nd Place: $10,000
3rd Place: $5,000
4th - 10th Places: $500 each
*** Dissemination of results ***
The top ranked participants will be invited to present their solutions at KDD2019, to be held in Anchorage, Alaska, from August 4 - 8, USA.
Additionally, participants will be invited to submit papers to a special issue on AutoML in a top tier journal (TBA)
*** Sponsors ***
This challenge is sponsored by 4Paradigm, Microsoft Research and ChaLearn
*** Organizing team ***
Wei-Wei Tu, 4Paradigm Inc., China - tuwe...@4paradigm.com (Main organizer)
Isabelle Guyon, Universté Paris-Saclay, France, ChaLearn, USA
Hugo Jair Escalante, CINVESTAV and INAOE, Mexico, ChaLearn, USA
Sergio Escalera, University of Barcelona, Spain
Evelyne Viegas, Microsoft Research
Mengshuo Wang, 4Paradigm Inc., China
Xiawei Guo, 4Paradigm Inc., China
Ling Yue, 4Paradigm Inc., China
Hai Wang, 4Paradigm Inc., China
Wenhao Li, 4Paradigm Inc., China
Yuanfei Luo, 4Paradigm Inc., China
Jian Liu, 4Paradigm Inc., China
Jingsong Wang, 4Paradigm Inc., China
Runxing Zhong, 4Paradigm Inc., China
Yadong Zhao, 4Paradigm Inc., China
Yuanmeng Huang, 4Paradigm Inc., China
Yuqiang Chen, 4Paradigm Inc., China
Wenyuan Dai, 4Paradigm Inc., China
Qiang Yang, Hong Kong University of Science and Technology, Hong Kong, China