Latent Ordinary Differential Equations for Irregularly-Sampled Time Series

Yulia Rubanova, Tian Qi Chen, David Duvenaud
2019 Neural Information Processing Systems  
Time series with non-uniform intervals occur in many applications, and are difficult to model using standard recurrent neural networks (RNNs). We generalize RNNs to have continuous-time hidden dynamics defined by ordinary differential equations (ODEs), a model we call ODE-RNNs. Furthermore, we use ODE-RNNs to replace the recognition network of the recently-proposed Latent ODE model. Both ODE-RNNs and Latent ODEs can naturally handle arbitrary time gaps between observations, and can explicitly
more » ... del the probability of observation times using Poisson processes. We show experimentally that these ODE-based models outperform their RNN-based counterparts on irregularly-sampled data.
dblp:conf/nips/RubanovaCD19 fatcat:ahq34rsdrnct7k2sxyn4yytlwe