Professor Aditya Grover and co-authors Noam Rozen, Maximilian Nickel, and Yaron Lipman from Weizmann Institute of Science and Meta AI, have received the Outstanding Paper Award at the Conference on Neural Information Processing Systems. NeurIPS 202 for their paper “Moser Flow: Divergence-based Generative Modeling on Manifolds.” Their paper builds on a classic result in differential geometry from Moser (1965) to propose a new family of continuous normalizing flow. generative models over Riemannian manifolds. The paper demonstrates state-of-the-art results in generative modeling of natural disasters over the earth and implicit surfaces with non-constant curvatures. NeurIPS is the premier conference for machine learning and artificial intelligence. Among the 9122 papers submitted to this conference this year, the committee selected only six papers for the Outstanding Paper Award.

Follow the link for the full award announcement: