New AI approach accelerates targeted materials discovery and paves the way for autonomous experiments

Scientists have developed an AI-powered method that can collect data more efficiently in the search for new materials, enabling researchers to tackle complex design challenges more accurately and quickly.

This research is a collaboration between computer science and materials science researchers at the Department of Energy’s SLAC National Accelerator Laboratory and Stanford University. The collaboration combines expertise in algorithm development, machine learning, and materials science.

Their work, published today in npj computer materialslays the foundation for “autonomous experiments,” where an intelligent algorithm sets the parameters for the next round of measurements at facilities such as SLAC’s Linac Coherent Light Source (LCLS). The new method also enables the rapid discovery of new materials, which could prove promising in areas such as climate change, quantum computing, and drug design.

Traditional materials discovery has always been a long and expensive process due to the cost of manufacturing and measuring the properties of new materials. The space of possible materials is also extremely large, exceeding 10 billion possibilities for materials made of just four elements. For pharmaceutical applications, the challenge is even greater, with around 10 billion possibilities.⁶⁰ potentially drug-like molecules containing only the most basic building blocks (C, H, O, N, and S atoms).

The task is further complicated by the need to meet complex design goals, such as discovering the conditions that allow nanoparticles of different sizes, shapes, and compositions to be synthesized. Traditional methods, which typically maximize or minimize a single property, are often too slow to search large search spaces to discover new materials that fit a researcher’s goals.

This paper proposes a new approach that takes complex objectives and automatically converts them into intelligent data acquisition strategies. One of the key features is the ability to learn and improve from each experiment, using AI to suggest next steps based on the data collected so far. The innovation is based on the concept of Bayesian Algorithm Execution (BAX), recently developed by co-author Willie Neiswanger, who was a postdoctoral researcher in computer science at Stanford when the research was conducted. In this method, a complex objective can be written as a simple shopping list or recipe, excelling in situations where multiple physical properties need to be considered.

Another important aspect is that this method is user-friendly and open source, allowing scientists around the world to use it and adapt it to their research, thus promoting collaboration and innovation on a global scale.

The researchers tested their approach on a variety of custom targets for nanomaterial synthesis and magnetic material characterization. The results showed that their methods were significantly more efficient than current techniques, especially in complex scenarios.

“Our method allows for complex objectives, which enables automatic optimization over a large design space, increasing the likelihood of finding amazing new materials,” said Sathya Chitturi, a doctoral student at SLAC and Stanford who led the research. “The Bayesian algorithm execution framework allows you to capture the intricacies of materials design tasks in an elegant and simple way.”

The ability to design materials with specific catalytic properties, for example, could improve chemical processes that lead to more efficient and sustainable ways of manufacturing goods and materials, reducing energy consumption and waste. In manufacturing, new materials could improve processes such as 3D printing, enabling more precise and sustainable production. In healthcare, tailored drug delivery systems can improve the targeting and release of therapeutics, improving their efficacy and reducing side effects.

Researchers are already implementing ways to integrate this framework into experimental and simulation-based research projects to demonstrate its broad applicability and effectiveness.

“This project is a great example of multidisciplinary collaboration between SLAC and Stanford,” said Daniel Ratner, SLAC’s Machine Learning Initiative leader. “Sathya has been able to adapt Willie’s fundamental research in algorithmic computing to solve real-world scientific problems in materials science.”

MLI researchers are now exploring applications for large-scale materials simulations, and Neiswanger, Ratner, and their collaborators have just published a related application of BAX to optimize the performance of particle accelerators.

“By combining advanced algorithms with targeted experimental strategies, our method facilitates and accelerates the process of discovering new materials,” said Chris Tassone, director of the Materials Science Division at the Stanford Synchrotron Radiation Lightsource (SSRL) at SLAC. “This can lead to new innovations and applications in many industries.”