Graph-Driven Generative Models for Heterogeneous Multi-Task Learning

TL;DR

This paper introduces a graph-driven generative model that unifies heterogeneous multi-task learning by leveraging shared graph structures, improving performance across healthcare applications such as clinical topic modeling and procedure recommendation.

Contribution

The paper presents a novel framework combining GCNs and variational autoencoders to jointly model multiple heterogeneous tasks sharing a common graph structure.

Findings

Outperforms existing state-of-the-art methods in healthcare tasks

Effectively leverages shared graph information across tasks

Boosts performance in clinical topic modeling and procedure recommendation

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.