Graph-Driven Generative Models for Heterogeneous Multi-Task Learning

Abstract

We propose a novel graph-driven generative model, that unifies multiple heterogeneous learning tasks into the same framework. The proposed model is based on the fact that heterogeneous learning tasks, which correspond to different generative processes, often rely on data with a shared graph structure. Accordingly, our model combines a graph convolutional network (GCN) with multiple variational autoencoders, thus embedding the nodes of the graph i.e., samples for the tasks) in a uniform manner while specializing their organization and usage to different tasks. With a focus on healthcare applications (tasks), including clinical topic modeling, procedure recommendation and admission-type prediction, we demonstrate that our method successfully leverages information across different tasks, boosting performance in all tasks and outperforming existing state-of-the-art approaches.

arXiv commentAccepted by AAAI-2020

References in corpus (14)

• Inductive Representation Learning on Large Graphs
• BPR: Bayesian Personalized Ranking from Implicit Feedback
• An Overview of Multi-Task Learning in Deep Neural Networks
• Discovering Discrete Latent Topics with Neural Variational Inference
• An Empirical Evaluation of Deep Learning for ICD-9 Code Assignment using MIMIC-III Clinical Notes
• Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processes
• Topic Compositional Neural Language Model
• A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks
• Zero-Shot Learning via Class-Conditioned Deep Generative Models
• Multi-Label Classification of Patient Notes a Case Study on ICD Code Assignment
• Kernel Topic Models
• An alternative text representation to TF-IDF and Bag-of-Words
• Improving Textual Network Learning with Variational Homophilic Embeddings
• Toward Incorporation of Relevant Documents in word2vec