Timezone: »

 
Poster
Equivariant Manifold Flows
Isay Katsman · Aaron Lou · Derek Lim · Qingxuan Jiang · Ser Nam Lim · Christopher De Sa

Tue Dec 07 04:30 PM -- 06:00 PM (PST) @
in Poster Session 2 »

Tractably modelling distributions over manifolds has long been an important goal in the natural sciences. Recent work has focused on developing general machine learning models to learn such distributions. However, for many applications these distributions must respect manifold symmetries—a trait which most previous models disregard. In this paper, we lay the theoretical foundations for learning symmetry-invariant distributions on arbitrary manifolds via equivariant manifold flows. We demonstrate the utility of our approach by learning quantum field theory-motivated invariant SU(n) densities and by correcting meteor impact dataset bias.

Keywords: Theory · Machine Learning · Generative Model

Author Information

Isay Katsman (Cornell University)
Aaron Lou (Cornell University)
Derek Lim (Massachusetts Institute of Technology)
Qingxuan Jiang (Massachusetts Institute of Technology)
Ser Nam Lim (Facebook AI)
Christopher De Sa (Cornell)

More from the Same Authors