Discovery and density estimation of latent confounders in Bayesian networks with evidence lower bound

Kiattikun Chobtham, Anthony C. Constantinou
Proceedings of The 11th International Conference on Probabilistic Graphical Models, PMLR 186:121-132, 2022.

Abstract

Discovering and parameterising latent confounders represent important and challenging problems in causal structure learning and density estimation respectively. In this paper, we focus on both discovering and learning the distribution of latent confounders. This task requires solutions that come from different areas of statistics and machine learning. We combine elements of variational Bayesian methods, expectation-maximisation, hill-climbing search, and structure learning under the assumption of causal insufficiency. We propose two learning strategies; one that maximises model selection accuracy, and another that improves computational efficiency in exchange for minor reductions in accuracy. The former strategy is suitable for small networks and the latter for moderate size networks. Both learning strategies perform well relative to existing solutions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v186-chobtham22a, title = {Discovery and density estimation of latent confounders in Bayesian networks with evidence lower bound}, author = {Chobtham, Kiattikun and Constantinou, Anthony C.}, booktitle = {Proceedings of The 11th International Conference on Probabilistic Graphical Models}, pages = {121--132}, year = {2022}, editor = {Salmerón, Antonio and Rumı́, Rafael}, volume = {186}, series = {Proceedings of Machine Learning Research}, month = {05--07 Oct}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v186/chobtham22a/chobtham22a.pdf}, url = {https://proceedings.mlr.press/v186/chobtham22a.html}, abstract = {Discovering and parameterising latent confounders represent important and challenging problems in causal structure learning and density estimation respectively. In this paper, we focus on both discovering and learning the distribution of latent confounders. This task requires solutions that come from different areas of statistics and machine learning. We combine elements of variational Bayesian methods, expectation-maximisation, hill-climbing search, and structure learning under the assumption of causal insufficiency. We propose two learning strategies; one that maximises model selection accuracy, and another that improves computational efficiency in exchange for minor reductions in accuracy. The former strategy is suitable for small networks and the latter for moderate size networks. Both learning strategies perform well relative to existing solutions. } }
Endnote
%0 Conference Paper %T Discovery and density estimation of latent confounders in Bayesian networks with evidence lower bound %A Kiattikun Chobtham %A Anthony C. Constantinou %B Proceedings of The 11th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2022 %E Antonio Salmerón %E Rafael Rumı́ %F pmlr-v186-chobtham22a %I PMLR %P 121--132 %U https://proceedings.mlr.press/v186/chobtham22a.html %V 186 %X Discovering and parameterising latent confounders represent important and challenging problems in causal structure learning and density estimation respectively. In this paper, we focus on both discovering and learning the distribution of latent confounders. This task requires solutions that come from different areas of statistics and machine learning. We combine elements of variational Bayesian methods, expectation-maximisation, hill-climbing search, and structure learning under the assumption of causal insufficiency. We propose two learning strategies; one that maximises model selection accuracy, and another that improves computational efficiency in exchange for minor reductions in accuracy. The former strategy is suitable for small networks and the latter for moderate size networks. Both learning strategies perform well relative to existing solutions.
APA
Chobtham, K. & Constantinou, A.C.. (2022). Discovery and density estimation of latent confounders in Bayesian networks with evidence lower bound. Proceedings of The 11th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 186:121-132 Available from https://proceedings.mlr.press/v186/chobtham22a.html.

Related Material