Advances in autonomous driving laboratories in the field of chemistry and materials science, utilizationArtificial intelligenceand automation, which is expected to revolutionize research by accelerating the discovery of new molecules and materials. Milad Abolhasani stressed the need to standardize definitions and performance metrics in order to effectively compare and improve these technologies.
The standardized metrics of the Autonomous Driving Lab are designed to accelerate discoveries in chemistry and materials science through collaborative improvements.
Interest in "autonomous labs" that use artificial intelligence (AI) and automated systems to accelerate research and discovery has surged in the fields of chemistry and materials science. Researchers have now come up with a set of definitions and performance metrics that will allow researchers, non-experts, and future users to better understand what these new technologies are doing and how each performs compared to other autonomous driving labs.
Autonomous driving labs hold great promise in accelerating the discovery of new molecules, materials, and manufacturing processes, with applications ranging from electronics to pharmaceuticals. While these technologies are still fairly new, some have been shown to reduce the time it takes to identify new materials from months or years to days.
Milad Abolhasani, associate professor of chemical and biomolecular engineering at North Carolina State University and corresponding author of an article on the new metric, said: "At the moment, the autonomous driving lab is attracting a lot of attention, but there are still a lot of open questions about these technologies. "This technology has been described as 'autonomous,' but different research teams define 'autonomous' differently. For the same reason, different research teams report different elements of their work in different ways. This makes it difficult to compare these technologies to each other, which is important if we want to be able to learn from each other and move the field forward.
What does Autonomous Lab A do well? "How can we use it to improve the performance of Autonomous Lab B? We have come up with a shared set of definitions and performance metrics that we hope everyone working in this field will be able to adopt. The ultimate goal will be for all of us to learn from each other and advance these powerful research-accelerating technologies.
"For example, we seem to be seeing some challenges in the autonomous driving lab related to the performance, precision and robustness of certain autonomous systems," Abolhasani said. "This raises the question of how useful these technologies really are. If we have standardized metrics and results reporting, we can identify these challenges and better understand how to address them. ”
At the heart of the new proposal is a clear definition of autonomous driving labs and seven proposed performance metrics that researchers will incorporate in any published work related to their autonomous driving labs.
Degree of autonomy: How much guidance does the system require from the user? Business lifespan: How long can a system run without user intervention? Throughput: How long does it take for the system to run an experiment? Experimental precision: How reproducible are the results of the system? Material usage: What is the total amount of material used by a system per experiment? Accessible parameter space: To what extent can the system account for all the variables in each experiment? Optimize efficiency.
Optimization efficiency is the most important of these metrics, but it's also the most complex – it doesn't fit into a succinct definition," Abolhasani said. "Essentially, we want researchers to quantitatively analyze the performance of their autonomous driving labs and their experimental selection algorithms by benchmarking baselines, such as random sampling.
"Ultimately, we believe that adopting a standardized approach to reporting on autonomous driving labs will help ensure that the field produces trustworthy, repeatable results, allowing AI programs to take full advantage of the large number of high-quality datasets generated by autonomous driving labs," Abolhasani said. ”