Improving Irregularly Sampled Time Series Learning with Dense Descriptors of Time
This addresses the problem of handling irregular time intervals in time series data for machine learning practitioners, but it is incremental as it builds on existing embedding and model approaches.
The paper tackled the challenge of supervised learning with irregularly sampled time series by proposing Time Embeddings, a method using sinusoidal functions to represent timestamps as dense vectors, which improved LSTM-based and classical models on MIMIC III dataset tasks, especially for very irregular data.
Supervised learning with irregularly sampled time series have been a challenge to Machine Learning methods due to the obstacle of dealing with irregular time intervals. Some papers introduced recently recurrent neural network models that deals with irregularity, but most of them rely on complex mechanisms to achieve a better performance. This work propose a novel method to represent timestamps (hours or dates) as dense vectors using sinusoidal functions, called Time Embeddings. As a data input method it and can be applied to most machine learning models. The method was evaluated with two predictive tasks from MIMIC III, a dataset of irregularly sampled time series of electronic health records. Our tests showed an improvement to LSTM-based and classical machine learning models, specially with very irregular data.