Past talk · AI/ML Seminar Series
Modeling Irregular Time Series with Continuous Recurrent Units
Senior Research Scientist, Bosch Center for AI
- Date & time
- Monday, February 7, 2022 · 1:00 PM
- Location
- Online (live stream)
Abstract
Recurrent neural networks (RNNs) are a popular choice for modeling sequential data but assume constant time-intervals between observations. In many datasets (e.g. medical records) observation times are irregular and can carry important information. We propose continuous recurrent units (CRUs) — a neural architecture that naturally handles irregular intervals. The CRU assumes a hidden state that evolves according to a linear stochastic differential equation, integrated into an encoder-decoder framework via the continuous-discrete Kalman filter in closed form. We find that the CRU can interpolate irregular time series better than methods based on neural ODEs.
About the speaker
Maja Rudolph is a Senior Research Scientist at the Bosch Center for AI, where she works on machine learning research derived from engineering problems such as modeling driving behavior and finding anomalies in sensor data. She completed her Ph.D. in Computer Science at Columbia University in 2018, advised by David Blei, and holds a BS in Mathematics from MIT.