GT MAP is a place for research discussion and collaboration. We welcome participation of any researcher interested in discussing his/her project and exchange ideas with Mathematicians. Mathematics can provide useful tools and insights to different research projects, and GT-MAP is to provide a channel of various discussions.
Feel free to consider giving presentations in our seminar series and workshop.
If you want to join GT-MAP mailing list, please add your email address to the subscribe button shown in the right bottom of this page.
If you are interested, here is a link to School of Mathematics weekly Seminars and Colloquia.
GT-MAP Seminar April 26th (Friday) 3PM
Speaker: Prof. Yao Xie – Industrial & Systems Engineering (ISyE) at Georgia Tech
Location: Skiles 005
Title: Computing High-Dimensional Optimal Transport by Flow Neural Networks
Abstract: Flow-based models are widely used in generative tasks, including normalizing flow, where a neural network transports from a data distribution P to a normal distribution. This work develops a flow-based model that transports from P to an arbitrary Q (which can be pre-determined or induced as the solution to an optimization problem), where both distributions are only accessible via finite samples. We propose to learn the dynamic optimal transport between P and Q by training a flow neural network. The model is trained to optimally find an invertible transport map between P and Q by minimizing the transport cost. The trained optimal transport flow subsequently allows for performing many downstream tasks, including infinitesimal density ratio estimation (DRE) and distribution interpolation in the latent space for generative models. The effectiveness of the proposed model on high-dimensional data is demonstrated by strong empirical performance on high-dimensional DRE, OT baselines, and image-to-image translation.