← All problems
🌍 IOAI 2025, Individual Day 2
Pixel Efficiency
Select a single small rectangle of each image (at most 6.25 percent of pixels) such that CLIP zero-shot classification still succeeds.
- Statement
- English · Original statement ↗
- Data
- Training data ↗ · Test data ↗ · Hugging Face dataset: IOAI-official/IOAI-2025-Pixel-ref ↗ · Hugging Face dataset: IOAI-official/IOAI2025 ↗
- To hand in
- A predictions file
- Hardware
- Needs a GPU (free Colab works)
- Data size
- 199.4 MB
- Data type
- 🖼️ Images
- Estimated difficulty
- ★★★★★
- Solution
- See a solution · best opened after you have tried
Get ready on the platform
The chapters of the learning path that prepare for this problem.
These chapters are picked from the solution: showing them may hint at the approach.
›How to tackle it
- Read the statement. What to hand in, the metric that scores you, the time and hardware limits.
- Explore the data. Size, format, missing values, how the labels are spread.
- Score yourself first. Keep part of the training data aside to measure your work before submitting.
- Start simple, then improve. A first solution that works end to end, then one change at a time.
- Compare with a solution. Once you have searched, open it and see what you would have done differently.
The statement, data and solutions belong to their authors; we only link to them. Licence : CC-BY-4.0.
Are you the author or organiser? Ask for a correction or removal