Exploiting machine learning to speed up the search for smarter energy storage
Batteries are to a clean-energy future what sunshine is to California: indispensable. But batteries today are weak, and so is the method that scientists are using to try to improve them. Batteries cost too much, hold too little energy, explode too easily, and lose their punch too fast. The search for better batteries, meanwhile, is largely a guessing game based on trial and error. But Stanford research is pointing up a new way to dramatically improve the quest for beefier batteries: reverse-engineering the scientific method through machine learning. This talk will explore that promise — a path for the world to develop the cutting-edge batteries that could facilitate a global energy revolution.
Austin Sendek | PhD student, Applied Physics