uGLAD: A deep learning model to recover conditional independence graphs

Harsh Shrivastava, Urszula Chajewska, Robin Abraham, Xinshi Chen

Keywords: Graphical Lasso, Deep Learning, Unrolled Algorithms, Conditional Independence graphs, Sparse graphs

TL;DR: An unsupervised deep learning method based on unrolled algorithm technique to recover conditional independence graphs.

Abstract: Probabilistic Graphical Models are generative models of complex systems. They rely on conditional independence assumptions between variables to learn sparse representations which can be visualized in a form of a graph. Such models are used for domain exploration and structure discovery in poorly understood domains. This work introduces a novel technique to perform sparse graph recovery by optimizing deep unrolled networks. Assuming that the input data $X\in\mathbb{R}^{M\times D}$ comes from an underlying multivariate Gaussian distribution, we apply a deep model on $X$ that outputs the precision matrix $\Theta$. Then, the partial correlation matrix \mathrm{P} is calculated which can also be interpreted as providing a list of conditional independence assertions holding in the input distribution. Our model, \texttt{uGLAD}, builds upon and extends the state-of-the-art model \texttt{GLAD} to the unsupervised setting. The key benefits of our model are (1) \texttt{uGLAD} automatically optimizes sparsity-related regularization parameters leading to better performance than existing algorithms. (2) We introduce multi-task learning based `consensus' strategy for robust handling of missing data in an unsupervised setting. We evaluate performance on synthetic Gaussian, non-Gaussian data generated from Gene Regulatory Networks, and present case studies in anaerobic digestion and infant mortality.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning