Courses
Courses offered by the group.
Equivariant Denoisers Cannot Copy Graphs: Align Your Graph Diffusion Models
We show that standard permutation equivariant denoisers cause severe limitations on such tasks, a problem that we pinpoint to their inability to break symmetries present in the noisy inputs.
AbODE: Ab initio Antibody Design using Conjoined ODEs
we propose a generative model for antibody design using conjoined interacting neural ODEs
Modular Flows: Differential Molecular Generation
We propose generative graph normalizing flow models, based on a system of coupled node ODEs, that repeatedly reconcile locally toward globally aligned densities for high quality molecular generation.
Provably expressive temporal graph networks
We analyze the representational power and limits of modern models for (event-based) temporal graphs. We leverage our theoretical insights to introduce an architecture that is provably more expressive than existing ones.
Why GANs are overkill for NLP
This work offers a novel theoretical perspective on why, despite numerous attempts, adversarial approaches to generative modeling …