Summary:
- Researchers have developed a machine learning system that can design new molecules with desired properties, speeding up the process of drug discovery and materials development.
- The system, called Chemputer, can explore a vast chemical space to identify promising candidate molecules, and then optimize the molecular structure to enhance the desired properties.
- This automated approach has the potential to revolutionize the way new materials and pharmaceuticals are developed, by making the process more efficient and cost-effective compared to traditional trial-and-error methods.