Improving Deep Learning With a Little Help From Physics
1 min read
Summary
Rose Yu is harnessing the power of artificial intelligence (AI) and physics to potentially revolutionise the way both disciplines are taught and understood.
As a child, Yu was given a computer, which sparked a lifetime passion for the machine.
Now, Yu is delving into the world of AI, melding it with her love of physics to create physics-guided deep learning.
Deep learning uses multiple layered neural networks to identify patterns in data, and Yu has used this to help with problems involving constantly moving things, such as predicting the flow of traffic.
She has also assisted scientists in understanding climate modelling and the prediction of hurricanes and other turbulent behaviour, using the Navier-Stokes equation.
Yu is currently working on a model which uses AI to aid scientists across a range of different tasks from data gathering to experiment design.
Yu envisages a future partnership between human researchers and AI, based on the principles of physics, which could enhance the discovery process.
