Checklist NEOAI 2026 Homework (take-home practice problems) · Task 1
Code Stars
Predict the final ranks of about 1,000 participants in each of 15 competitive-programming contests from their historical results.
The task
The data are historical results of competitive-programming contests on a Codeforces-style platform: per-contest ranks, points, penalties and hacks for anonymised participants; per-problem results; problem metadata (maximum points, difficulty rating, tags); and contest metadata (division, scoring type, duration, start time).
For each of the 15 test contests, a list of about 1,000 participants is given and the contestant must reconstruct the order of the final leaderboard by predicting each participant's rank.
Abridged by SOTA from the official materials. The official statement has the exact rules, and it wins wherever this summary differs.
At a glance
- You get
train.parquet,train_results.parquet,problems.parquet,contest.parquet,test.parquet(contestId, handle),baseline_submission.csv.- You submit
- CSV with columns Id (contestId_handle) and rank (float or integer).
- Scoring
- Mean over test contests of 0.3·Recall@100 + 0.7·τ, where Recall@100 is the share of the true top 100 placed in the predicted top 100 and τ is the Kendall rank correlation over all participants of the contest.
- Rules
- Pretrained models and libraries outside the standard Kaggle environment are prohibited.
- Format
- Take-home practice problem for NEOAI 2026 participants (Kaggle, 12 Apr – 1 May 2026). Individual.