Applications of Machine Learning in the Mathematical Physics of Polymers
Time: 2019-01-08
Published By: Xiaoni Tan
Speaker(s): Jeff Z. Y. Chen (University of Waterloo)
Time: 10:30-11:30 January 8, 2019
Venue: Room 77201, Jingchunyuan 78, BICMR
A feed-forward neural network has a remarkable property which allows the network itself to be a universal approximator for any functions. Here we present a universal, machine-learning based solver for multi-variable partial differential equations. The algorithm approximates the target functions by neural networks and adjusts the network parameters to approximate the desirable solutions. The idea can be easily adopted for dealing with multi-variable, coupled integrodifferential equations, such as those in the self-consistent field theory of predicting polymer microphase- separated structures.
