Machine learning

A*STAR researchers develop machine learning models that quickly and accurately simulate chemical reactions involving complex liquid interactions.

Advanced photonic processors boost machine learning efficiency by handling vast data streams in parallel, with potential benefits for enhanced patient monitoring and cloud computing.

Researchers find that pre-training AI models on a global scale boosts their performance in machine learning tasks, particularly in scenarios with limited training data.

A robust machine learning approach to predict and optimise fuel consumption in ships is critical to practically enhance maritime environmental sustainability.

The innovative integration of robotics and machine learning accelerates the development of efficient, sustainable hydrogen catalysts.

Smartwatch data points to a connection between physical activity and prolonged COVID-19 symptoms among healthcare workers.

Aided by machine learning, researchers develop a novel high-performance steel optimised for more eco-friendly 3D printing.

A new machine learning method enables smaller models to understand and replicate complex relationships within large networks.

An enhanced computational design method uses deep learning to accelerate the development of light-based computer chip technologies.

A new algorithm accurately predicts when industrial machines will need maintenance, which can greatly improve efficiency and reduce costs.

A*STAR scientists use advanced computer models to create a massive database of natural products for drug discovery and beyond.

A new machine learning framework helps train robots to safely navigate pedestrians in crowded, real-world settings.