Our paper "Generative Causal Representation Learning for Out-of-Distribution Motion Forecasting" just got accepted to #ICML2023!
Congrats to my PhD student Shayan as well as postdoc
Zahra Gharaee for all their hard work on this collab with Oliver Schulte.
Arxiv link: https://arxiv.org/abs/2302.08635
We propose a new framework for leveraging causal information to improve robustness of learning in the presence of distribution shift. Empirical results on the motion forecasting domain support the theoretical findings.