Machine Learning Model Developed to Accelerate Discovery of Sustainable, Biodegradable Plastic Alternatives
-
Researchers developed a machine learning model to discover biodegradable plastic alternatives faster and more effectively.
-
The model is trained on a library of nanocomposite films made from natural materials, prepared by an automated robot.
-
Through iterative active learning, the model learns to predict material properties based on composition.
-
This approach reduces time and resources compared to traditional trial-and-error methods.
-
The researchers plan to expand the range of sustainable materials and applications to reduce plastic pollution.