Task-wise Split Gradient Boosting Trees for Multi-center Diabetes Prediction (2021-08-14T00:00:00.000000Z)

TL;DR

The proposed Task-wise Split Gradient Boosting Trees (TSGB) is proposed for the multi-center diabetes prediction task and reveals a problem when directly applying GBDT in MTL, i.e., the negative task gain problem.

Abstract

Diabetes prediction is an important data science application in the social healthcare domain. There exist two main challenges in the diabetes prediction task: data heterogeneity since demographic and metabolic data are of different types, data insufficiency since the number of diabetes cases in a single medical center is usually limited. To tackle the above challenges, we employ gradient boosting decision trees (GBDT) to handle data heterogeneity and introduce multi-task learning (MTL) to solve data insufficiency. To this end, Task-wise Split Gradient Boosting Trees (TSGB) is proposed for the multi-center diabetes prediction task. Specifically, we firstly introduce task gain to evaluate each task separately during tree construction, with a theoretical analysis of GBDT's learning objective. Secondly, we reveal a problem when directly applying GBDT in MTL, i.e., the negative task gain problem. Finally, we propose a novel split method for GBDT in MTL based on the task gain statistics, named task-wise split, as an alternative to standard feature-wise split to overcome the mentioned negative task gain problem. Extensive experiments on a large-scale real-world diabetes dataset and a commonly used benchmark dataset demonstrate TSGB achieves superior performance against several state-of-the-art methods. Detailed case studies further support our analysis of negative task gain problems and provide insightful findings. The proposed TSGB method has been deployed as an online diabetes risk assessment software for early diagnosis.

Authors

References33 items

Product-Based Neural Networks for User Response Prediction over Multi-Field Categorical Data

A Survey on Multi-Task Learning

A decision tree framework for understanding blast-induced mild Traumatic Brain Injury in a military medical database

A Clinical Decision Tree to Predict Whether a Bacteremic Patient Is Infected With an Extended-Spectrum β-Lactamase-Producing Organism.

XGBoost: A Scalable Tree Boosting System

Heterogeneous Multi-task Semantic Feature Learning for Classification

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

On handling negative transfer and imbalanced distributions in multiple source transfer learning

Tree-Based Ensemble Multi-Task Learning Method for Classification and Regression

Nonlaboratory-Based Risk Assessment Algorithm for Undiagnosed Type 2 Diabetes Developed on a Nation-Wide Diabetes Survey

Learning Multiple Tasks with Boosted Decision Trees

Sparse coding for multitask and transfer learning

Learning Task Grouping and Overlap in Multi-task Learning

Marginalized Denoising Autoencoders for Domain Adaptation

Clustered Multi-Task Learning Via Alternating Structure Optimization

Learning Nonlinear Functions Using Regularized Greedy Forest

Scikit-learn: Machine Learning in Python

Multi-task learning for boosting with application to web search ranking

An accelerated gradient method for trace norm minimization

DTU: A Decision Tree for Uncertain Data

Convex multi-task feature learning

Biographies, Bollywood, Boom-boxes and Blenders: Domain Adaptation for Sentiment Classification

A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data

Task-wise Split Gradient Boosting Trees for Multi-center Diabetes Prediction