Research

My research is all about analyzing the relationships between systems. How does a given system evolve in time? Is information about one system useful for predicting another one? Is there a cause-effect relationship between these systems? Most importantly, can I come up with a computer program that will answer these questions using the available data?

My main research project, with my advisor Petar Djuric and our collaborators, studies these problems in the context of multivariate time series. We also apply these models to neuroscience, where our goal is to improve medical diagnostics and further our anatomical knowledge of the brain.

Research interests:

Bayesian machine learning
Causality
Dynamical systems
Explainable AI
Time series modeling and prediction
Statistical signal processing

Keywords: Gaussian processes, causality, manifold, time series, differentiation, state space reconstruction

A plot showing a shadow manifold embedded in 3 dimensional space.

Publications

Here is a running list of publications and links to their associated materials.

Journal

Measuring Strength of Joint Causal Effects
Kurt Butler, Guanchao Feng, and Petar M. Djuric
IEEE Transactions on Signal Processing, 2024
[paper][Github]
Explainable Learning with Gaussian Processes
Kurt Butler, Guanchao Feng, and Petar M. Djuric
arXiv preprint, 2024.
[arXiv] [Github]
Dagma-DCE: Interpretable, Non-Parametric Differentiable Causal Discovery
Daniel Waxman, Kurt Butler, and Petar M. Djuric
IEEE Open Journal of Signal Processing, 2024
[paper] [arXiv] [Github]
On Causal Discovery with Convergent Cross Mapping
Kurt Butler, Guanchao Feng, and Petar M. Djuric
IEEE Transactions on Signal Processing, 2023
[paper] [Github]
A Differential Measure of the Strength of Causation
Kurt Butler, Guanchao Feng, and Petar M. Djuric
IEEE Signal Processing Letters, 2022
[paper] [Github]

Conference

Learning the hierarchical organization of the frontal lobe with differential causal effects
Kurt Butler, Duncan Cleveland, Charles B. Mikell, Sima Mofakham, Yuri B. Saalmann, and Petar M. Djuric
Science Talks, 2024
[video]
Sequential Detection of Anomalies in Noisy Outputs of an Unknown Function using Gaussian and Yule-Simon Processes
Liu Yang, Kurt Butler, and Petar M. Djuric
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
[paper]
An Approach to Learning the Hierarchical Organization of the Frontal Lobe
Kurt Butler, Duncan Cleveland, Charles B. Mikell, Sima Mofakham, Yuri B. Saalmann, and Petar M. Djuric
Proceedings of the 31st European Signal Processing Conference (EUSIPCO), 2023
[paper]
Detecting Confounders in Multivariate Time Series using Strength of Causation
Yuhao Liu, Chen Cui, Daniel Waxman, Kurt Butler, and Petar M. Djuric
Proceedings of the 31st European Signal Processing Conference (EUSIPCO), 2023
[paper]
Predicting latent states of dynamical systems with state-space reconstruction and Gaussian processes
Kurt Butler, Guanchao Feng, Charles B. Mikell, Sima Mofakham, and Petar M. Djuric
Proceedings of the 30th European Signal Processing Conference (EUSIPCO), 2022
[paper]