UNIST ECE student, (Anh Tong), and Prof. Jaesik Choi’s paper has been accepted at ICML 2019.
– Authors : Anh Tong, Jaesik Choi (corresponding author)
– Title: Discovering Latent Covariance Structures for Multiple Time Series
– Conference : International Conference on Machine Learning (ICML 2019)
A research paper entitled “Discovering Latent Covariance Structures for Multiple Time Series” has been accepted to ICML(International Conference on Machine Learning) 2019. The paper is authored by Anh Tong and Prof. Jaesik Choi. The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning.
The paper proposes a way to understand the relations among multiple time series by jointly learning structural Gaussian Process covariances and Indian Buffet Process latent variables. The model outputs better predictions on real-world data sets. Additionally, new types of explainable reports are introduced as a tool to compare time series pairwisely, giving more insight about data.