¹úÄ£¸£ÀûÔÚÏß

The Australian Institute for Machine Learning (AIML) Summer Grand Discovery Challenge is a machine learning competition designed to solve real-world problems using applied AI, engaging the talents and skills of undergraduate and postgraduate students at Adelaide University. 

The inaugural AIML Summer Grand Discovery Challenge is inviting interdisciplinary student teams to apply machine learning to real datasets provided by researchers within the University, industry or government, to solve specific data-driven problems. Students will work in teams to develop predictive models, collaborating closely with leading researchers and experts in machine learning. The challenge offers a chance to build their portfolio, showcase their skills, and compete for recognition and potential future research opportunities.

Key dates for 2025-26

Competition launch: Tuesday, 28 October 2025

Team registration deadline: Friday, 14 November 2025

Competition closes: Friday, 14 February 2026 (11:59 PM ACDT)

Winner announcement: Monday, 16 February 2026

Awards ceremony: Friday, 20 February 2026, 4:00 PM at AIML

Competition Overview

The competition will provide a problem statement and a dataset, typically split into training data (which includes labels) and test data (which does not). For 2025–26, the challenge focuses on predicting lettuce dry shoot weight from RGB and depth images.

Students will:

• Build regression models to predict DryWeightShoot (g/plant) from provided imagery.
• Use supplied datasets and preprocessing pipelines to ensure consistent results.

• Submit predictions, technical reports, and reproducible code.

The use of external data or public benchmarks is not permitted for this competition. Please train your model only using the dataset provided by the competition, as no external datasets are allowed. Each team may only make one daily submission of original work. Competition organisers reserve the right to disqualify teams for any misconduct. 

Forming a team

• The AIML Summer Grand Discovery Challenge is open to students at Adelaide University (including the University of Adelaide and the University of South Australia) enrolled in an undergraduate or postgraduate degree. 

• There is a maximum of 3 members per team. Interdisciplinary participation is welcome and encouraged. 

• Each student may only join one team.

Submission

The dataset originates from the 3rd International Autonomous Greenhouse Challenge (WUR, 2021).

Inputs:
  - RGB images (1920×1080, PNG).
  - Depth images (16-bit PNG, aligned to RGB).
Target: DryWeightShoot (grams).
Train/Test Split:
  - Train: 230 samples.
  - Test: 157 samples.
Important: Auxiliary traits (height, diameter, fresh weight, leaf area) may be used during training but will not be provided in test data.

Please submit your training script and we will run the training script to ensure that your model can reproduce the prediction results. Please make sure you provide all the details about how to run your training script.

Submission limits: 5 per team per day. Late submissions not accepted.

Teams must submit:
1. Prediction file (CSV) following template format.
2. Technical report (max 4 pages) including problem understanding, preprocessing, model design, and results.
3. Source code (Jupyter notebook or Python scripts) with documentation and reproducibility.
   - Include requirements/environment file.
   - Ensure compatibility with the preprocessing pipeline.

Evaluation and scoring

Primary metric: Mean Absolute Error (MAE) on DryWeightShoot.
Judging criteria:

• Model performance (60%)
• Code quality & reproducibility (20%)
• Technical report (15%)
• Innovation & relevance (5%)

Public leaderboard (30% of test data), private leaderboard (70% of test data). Final ranks are based on private scores.

Prizes

Prizes will be awarded separately for two team categories:

• HDR Teams: Teams that include one or more Higher Degree by Research (HDR) students.
• Undergraduate Teams: Teams composed entirely of undergraduate students.

Prizes will be distributed as follows:

• 1st Prize (HDR Category): $3,000 – awarded to the top-ranked team including HDR students.
• 1st Prize (Undergraduate Category): $3,000 – awarded to the top-ranked undergraduate team.
• Runner-Up (HDR Category): $1,000 – awarded to the second-ranked HDR team.
• Runner-Up (Undergraduate Category): $1,000 – awarded to the second-ranked undergraduate team.

Certificates will be given to all teams. The winning teams will present their results at an awards ceremony and may be featured in AIML promotional material.

Resources and support

The AIML Summer Grand Discovery Challenge is modelled on Kaggle competitions, online competitions where participants build and submit machine learning models to solve specific data-driven problems using provided datasets. Visit the for more information on this format.

AIML staff are also available to provide technical support on the competition platform. AIML researchers are available for consultation during office hours. Preprocessing guidance is provided in the supplied scripts for the competition.