NeurIPS - Open Polymer Prediction 2025

Overview

This hackathon challenges you to predict the properties of polymers using machine learning, aiming to speed up the discovery of new sustainable materials. You'll use chemical structures provided as text strings to forecast key metrics like glass transition temperature and molecular size. Your work can help researchers develop eco-friendly materials faster and make a positive impact on the planet.

Requirements

Teams can have any number of members, but the deadline to join or merge teams is September 8, 2025. All submissions must be made through Kaggle Notebooks. Your Notebook must run within 9 hours on either CPU or GPU and have internet access disabled. You are allowed to use freely available external data and pre-trained models. Your final submission needs to be a CSV file named 'submission.csv' with 'id' and predictions for five polymer properties: Tg, FFV, Tc, Density, and Rg.

Prizes

There's a total prize pool of $50,000, with cash prizes for the top five teams: 1st place gets $12,000, 2nd place gets $10,000, 3rd place gets $10,000, 4th place receives $8,000, and 5th place wins $5,000. Additionally, the top student team will receive an extra $5,000.