The 3rd Apache Flink Geek Challenge and AAIG CUP registration begins!

Introduction to The 3rd Geek Challenge, which is highly anticipated, registration is in progress~

With the impact of massive data, the value of data processing and analysis capabilities in the business is increasing day by day, and the exploration of the timeliness of data processing in all walks of life is also deepening, as the main real-time computing computing engine-Apache Flink came into being.

In order to bring more real-time computing empowerment practice ideas to the industry and encourage developers who love technology to deepen their grasp of Flink, the Apache Flink community has teamed up with Alibaba Cloud, Intel, and Alibaba Artificial Intelligence Governance and Sustainability Laboratory (AAIG). ), Occlum jointly organized the "The 3rd Apache Flink Geek Challenge and AAIG CUP" activity, officially launched today.

highlights: as the world's first real-time traffic attack detection competition based on a recommendation system, we released the real data set of 1 million traffic cheating identification for the first time! And apply Intel SGX encryption technology to ensure the data security of the risk control model.

👉 Click for more event information 👈

In the past two years, Alibaba Cloud, Intel, and the Apache Flink community have jointly held the first and second Flink Geek Challenge, using platforms and technologies such as Flink, Analytics Zoo, and Proxima to guide players in garbage classification, real-time epidemic tracking, etc. Thinking and innovating on hot issues of people’s livelihood; Alibaba Artificial Intelligence Governance and Sustainability Laboratory (AAIG) hosted seven security AI challenger programs, such as face recognition contest, abusive text contest, defense model white box and no Limit competitions, etc., bravely break into AI to fight no man's land, and launch a new project AAIG-CUP that combines AI governance with multiple disciplines.

This year's 3rd Flink Geek Challenge Contest and AAIG CUP utilizes the Intel® SGX technology of the third-generation Intel® Xeon® Scalable processor to guide players in the exploration and development of privacy-protected machine learning. This competition will continue to be open to the whole society, and individuals, universities, research institutes, enterprises, and maker teams can register for the competition.

Competition background

With the development of the Internet, online shopping has become the choice of more and more people. According to Alibaba's financial report, the total turnover of Alibaba's website in fiscal year 2020 will exceed one trillion US dollars, and the global annual active consumers will reach 960 million.

In order to meet the individual needs of different users, the e-commerce platform will recommend suitable products according to the user's hobbies, so as to realize the needs of thousands of people in ordering products. In order to obtain more platform traffic exposure and expose their products to more consumers, some merchants increase the exposure opportunities of their products through the recommendation mechanism of the HACK platform. One of the typical tactics is the "thigh hug" attack. This method employs a group of malicious users to click on target products and popular products to establish the relationship between the target products and popular products, and improve the target products and popular products. I2I correlation points between products.

In this way, merchants induce users to purchase products that do not match their names with the psychological anticipation of exploding money, which not only harms the interests of consumers, reduces their shopping experience, but also affects the reputation of the platform and other merchants, and seriously disrupts the fairness of the platform . Real-time interception of such behaviors helps protect the real-time recommendation system from malicious attacks while ensuring the timeliness of recommendations.

How to accurately and efficiently identify this type of malicious traffic attack and filter malicious click data in real time is an urgent problem in the recommendation system.

In addition, such real-time risk control systems have higher requirements for data security. If the system's interception algorithm is accidentally leaked, the HACK platform will be able to strengthen the camouflage capabilities of malicious traffic and increase the difficulty of the platform to monitor malicious traffic. Therefore, such systems must be deployed in an encrypted and trusted environment.

This contest requires players to build a PPML (Privacy Preserving Machine Learning) application to protect data security based on Flink, Analytics Zoo/BigDL and other components in the Occlum environment to realize real-time identification of malicious traffic.

Schedule

The competition is divided into three stages: preliminary, semi-finals and finals. The schedule is as follows:

Preliminary stage

The participating teams will use the docker image submission method, and the contestants will submit the packaged code image to run to get the build result; the participating teams will download small training sets and test sets through the Tianchi platform for local training and debugging algorithms; the final completed code After the image is submitted to the Tianchi platform, it is trained and predicted by the evaluation system on the complete data set.

rematch stage

After entering the semi-finals, players will be further required to train deep learning models on semi-supervised data sets, and they will be encouraged to do some optimization and innovation in engineering.

final stage

The final finals will be held in the form of a defense meeting. The teams that qualify for the finals need to prepare the defense PPT in advance as required, and the on-site judges will make a comprehensive score based on the players' technical ideas and theoretical depth on-site performance.

Activity incentive

It is reported that this competition will eventually produce:

• champion team with a prize of 100,000 RMB;
• runners-up teams 2 teams, each with a prize of 50,000 yuan;
• third teams 3 teams, each with a prize of 20,000 yuan;
• Winning Award 4 teams with a prize of RMB 10,000.

At the same time, this competition will continue the "Encouragement Program" gameplay during the registration stage. Successfully inviting friends to participate can become an encouragement teacher to win a generous gift!

eggs: The award ceremony of this competition is expected to be held in the main forum of Flink Forward Asia 2021 (recently online) in December. For event information, please pay attention to the "Flink Chinese Community" official account, and more tips for participating in the competition are waiting for you to get~!

👉 Click for more event information 👈

Copyright Statement: content of this article is contributed spontaneously by Alibaba Cloud real-name registered users, and the copyright belongs to the original author. The Alibaba Cloud Developer Community does not own its copyright and does not assume corresponding legal responsibilities. For specific rules, please refer to the "Alibaba Cloud Developer Community User Service Agreement" and the "Alibaba Cloud Developer Community Intellectual Property Protection Guidelines". If you find suspected plagiarism in this community, fill in the infringement complaint form to report it. Once verified, the community will immediately delete the suspected infringing content.