Relate and Predict: Structure-Aware Prediction with Jointly Optimized Neural DAG

TL;DR

This paper introduces dGAP, a neural network framework that learns feature dependency structures and makes predictions simultaneously, resulting in more accurate and interpretable models that reveal feature relationships.

Contribution

The paper presents a novel joint optimization approach for structure-aware prediction that explicitly models feature dependencies within a neural network framework.

Findings

dGAP outperforms existing methods in prediction accuracy.

dGAP successfully recovers true feature dependency structures.

The model provides interpretable insights into feature relevance.

Abstract

Understanding relationships between feature variables is one important way humans use to make decisions. However, state-of-the-art deep learning studies either focus on task-agnostic statistical dependency learning or do not model explicit feature dependencies during prediction. We propose a deep neural network framework, dGAP, to learn neural dependency Graph and optimize structure-Aware target Prediction simultaneously. dGAP trains towards a structure self-supervision loss and a target prediction loss jointly. Our method leads to an interpretable model that can disentangle sparse feature relationships, informing the user how relevant dependencies impact the target task. We empirically evaluate dGAP on multiple simulated and real datasets. dGAP is not only more accurate, but can also recover correct dependency structure.