Artificial Intelligence 5 min read

58 City AI Algorithm Competition – Updates, Leaderboard, and Baseline Model

The 58 City AI Algorithm Competition, now in its second edition, has attracted over 200 teams from more than 50 universities and 30 internal groups, with daily leaderboard updates, a baseline MMoE model scoring 0.7294, and a GitHub repository for participants.

58 Tech
58 Tech
58 Tech
58 City AI Algorithm Competition – Updates, Leaderboard, and Baseline Model

The 58 City AI Algorithm Competition (second edition) launched on July 9, 2021, drawing over 200 teams, including corporate non‑students and university students.

Participants come from more than 50 domestic universities such as University of Science and Technology of China, University of Chinese Academy of Sciences, Xi'an Jiaotong University, Harbin Institute of Technology, Huazhong University of Science and Technology, Beihang University, University of Electronic Science and Technology of China, and Chongqing University of Posts and Telecommunications.

The event is organized by the 58 Technology Committee, co‑hosted by TEG‑AI Lab and HRG‑User Side Technology Department, with over 30 internal 58 Group teams also competing.

For detailed information, see the official announcement link.

Current teams are continuously submitting results, and the competition is in full swing with daily leaderboard updates.

01 Daily Top Team – Since the leaderboard opened on July 23, the team “word很大” led for two days, followed by “摆烂ing” and “下次一定队” each holding the top spot for three days. Subsequent changes saw “乘风破浪”, “下次一定队”, “无趣的人”, and “word很大” alternating at the top, with “无趣的人” currently leading again.

02 Current Top Ten – On August 2, 2021, at 10:02 AM, the team “下次一定队” from 58 City broke the 0.8‑point threshold and ranked first, but was overtaken six hours later. Later that day, “无趣的人” from Chongqing University reached first place at 4:04 PM, pushing “下次一定队” to second. On August 3, “word很大” from Chongqing University of Posts and Telecommunications rose to second place at 10:01 AM, moving “下次一定队” to third. On August 4, “word很大” narrowly led the board at 3:51 PM, with “无趣的人” second and “下次一定队” third. On August 5, “无趣的人” reclaimed the top spot at 2:23 PM, followed by “word很大” and “下次一定队”. The top‑ten list as of 15:49 on August 5 is shown in the accompanying image.

03 Submission Status – The team “Dream” from University of Electronic Science and Technology of China has submitted 37 times, the most active participant. Even at 4:03 AM, “Dream” was still submitting, while “数据魔术师” from Huazhong University of Science and Technology was working at 3:26 AM, illustrating the dedication of the competitors.

04 Friendly Reminder – Submissions close on August 23, 2021, leaving 20 days remaining; participants are encouraged to submit promptly.

Baseline Model – A baseline model using the MMoE architecture currently scores 0.7294. The accompanying article explains how to reproduce the baseline results within half an hour. The baseline code is open‑sourced at https://github.com/wuba/dl_inference/tree/master/DLPredictOnline/demo/model/pytorch/mmoe .

05 Contact – To join the competition discussion group, search for “jishu-58” on WeChat and add the 58 technical assistant.

machine learningAI competitionMMoEleaderboard58 Citybaseline model
58 Tech
Written by

58 Tech

Official tech channel of 58, a platform for tech innovation, sharing, and communication.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.