COMPUTATIONAL AND DATA-DRIVEN SOLUTIONS FOR THE PHYSICAL SCIENCES
We’re a group of researchers at the interface of science, statistics and computation driven to understand how to best help researchers discover new science through the use of computers, data analysis, machine learning and decision theory.
Our work applies the core capabilities of Princeton’s ORFE department, OptLab and Castle Labs towards problems in Materials Science and Chemistry to deliver “materials-genomic” solutions.
Our approach is a synergistic combination of physics-based modeling with statistical techniques to perform “physics-aware” machine learning, with the goal of accelerated materials discovery.
Our main thrust is in efficient guided experiments using Optimal Learning.
Drawing from the expertise of Princeton’s ORFE department, OptLab and Castle Labs, our core group of researchers span a diverse set of fields, including operations research, statistics, electrical engineering, applied mathematics and materials science.
WE’RE HERE TO HELP
We recognize that, when it comes to scientific research, there isn’t a one-size-fits-all solution to your computational and data needs. That’s why each of our tailored solutions are a result of an active collaboration with scientists, incorporating their domain knowledge, assumptions and existing data.
Here are just a few examples of how scientists can leverage our expertise:
We’ll work with you in developing and executing a novel computational or data-driven program that complements and accelerates your research.
Run and visualize simulations
Utilize our computational resources to run simulations without the hassle of setting up and maintaining code yourself. Learn more »
Explore experiment space
Couple decision theory with physics to efficiently explore combinatorial experiment space. Learn more »
Use machine learning to make property predictions based on your data and literature review. Learn more »