Inferring the time-varying coupling of dynamical systems with temporal convolutional autoencoders
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909217371193344 |
|---|---|
| author | Calderon, Josuan Berman, Gordon J. |
| author_facet | Calderon, Josuan Berman, Gordon J. |
| contents | Most approaches for assessing causality in complex dynamical systems fail when the interactions between variables are inherently non-linear and non-stationary. Here we introduce Temporal Autoencoders for Causal Inference (TACI), a methodology that combines a new surrogate data metric for assessing causal interactions with a novel two-headed machine learning architecture to identify and measure the direction and strength of time-varying causal interactions. Through tests on both synthetic and real-world datasets, we demonstrate TACI's ability to accurately quantify dynamic causal interactions across a variety of systems. Our findings display the method's effectiveness compared to existing approaches and also highlight our approach's potential to build a deeper understanding of the mechanisms that underlie time-varying interactions in physical and biological systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_03212 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Inferring the time-varying coupling of dynamical systems with temporal convolutional autoencoders Calderon, Josuan Berman, Gordon J. Machine Learning Quantitative Methods Most approaches for assessing causality in complex dynamical systems fail when the interactions between variables are inherently non-linear and non-stationary. Here we introduce Temporal Autoencoders for Causal Inference (TACI), a methodology that combines a new surrogate data metric for assessing causal interactions with a novel two-headed machine learning architecture to identify and measure the direction and strength of time-varying causal interactions. Through tests on both synthetic and real-world datasets, we demonstrate TACI's ability to accurately quantify dynamic causal interactions across a variety of systems. Our findings display the method's effectiveness compared to existing approaches and also highlight our approach's potential to build a deeper understanding of the mechanisms that underlie time-varying interactions in physical and biological systems. |
| title | Inferring the time-varying coupling of dynamical systems with temporal convolutional autoencoders |
| topic | Machine Learning Quantitative Methods |
| url | https://arxiv.org/abs/2406.03212 |