PyTorch implementation of the NCDSSM models presented in the ICML '23 paper "Neural Continuous-Discrete State Space Models for Irregularly-Sampled Time Series".
Hi, I'm very excited about the emerge of this work. It makes sense that those traditional deep models all have high probability of converting themself from discrete to continuous form. However, it costs much time to deeply understand the paper without review on openreview.com. Could you provide the review to me? Thank you~