IDYNO: Learning Nonparametric DAGs from Interventional Dynamic Data

Tian Gao, Debarun Bhattacharjya, Elliot Nelson, Miao Liu, Yue Yu
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:6988-7001, 2022.

Abstract

Causal discovery in the form of a directed acyclic graph (DAG) for time series data has been widely studied in various domains. The resulting DAG typically represents a dynamic Bayesian network (DBN), capturing both the instantaneous and time-delayed relationships among variables of interest. We propose a new algorithm, IDYNO, to learn the DAG structure from potentially nonlinear times series data by using a continuous optimization framework that includes a recent formulation for continuous acyclicity constraint. The proposed algorithm is designed to handle both observational and interventional time series data. We demonstrate the promising performance of our method on synthetic benchmark datasets against state-of-the-art baselines. In addition, we show that the proposed method can more accurately learn the underlying structure of a sequential decision model, such as a Markov decision process, with a fixed policy in typical continuous control tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-gao22a, title = {{IDYNO}: Learning Nonparametric {DAG}s from Interventional Dynamic Data}, author = {Gao, Tian and Bhattacharjya, Debarun and Nelson, Elliot and Liu, Miao and Yu, Yue}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {6988--7001}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/gao22a/gao22a.pdf}, url = {https://proceedings.mlr.press/v162/gao22a.html}, abstract = {Causal discovery in the form of a directed acyclic graph (DAG) for time series data has been widely studied in various domains. The resulting DAG typically represents a dynamic Bayesian network (DBN), capturing both the instantaneous and time-delayed relationships among variables of interest. We propose a new algorithm, IDYNO, to learn the DAG structure from potentially nonlinear times series data by using a continuous optimization framework that includes a recent formulation for continuous acyclicity constraint. The proposed algorithm is designed to handle both observational and interventional time series data. We demonstrate the promising performance of our method on synthetic benchmark datasets against state-of-the-art baselines. In addition, we show that the proposed method can more accurately learn the underlying structure of a sequential decision model, such as a Markov decision process, with a fixed policy in typical continuous control tasks.} }
Endnote
%0 Conference Paper %T IDYNO: Learning Nonparametric DAGs from Interventional Dynamic Data %A Tian Gao %A Debarun Bhattacharjya %A Elliot Nelson %A Miao Liu %A Yue Yu %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-gao22a %I PMLR %P 6988--7001 %U https://proceedings.mlr.press/v162/gao22a.html %V 162 %X Causal discovery in the form of a directed acyclic graph (DAG) for time series data has been widely studied in various domains. The resulting DAG typically represents a dynamic Bayesian network (DBN), capturing both the instantaneous and time-delayed relationships among variables of interest. We propose a new algorithm, IDYNO, to learn the DAG structure from potentially nonlinear times series data by using a continuous optimization framework that includes a recent formulation for continuous acyclicity constraint. The proposed algorithm is designed to handle both observational and interventional time series data. We demonstrate the promising performance of our method on synthetic benchmark datasets against state-of-the-art baselines. In addition, we show that the proposed method can more accurately learn the underlying structure of a sequential decision model, such as a Markov decision process, with a fixed policy in typical continuous control tasks.
APA
Gao, T., Bhattacharjya, D., Nelson, E., Liu, M. & Yu, Y.. (2022). IDYNO: Learning Nonparametric DAGs from Interventional Dynamic Data. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:6988-7001 Available from https://proceedings.mlr.press/v162/gao22a.html.

Related Material