Learning Tasks for Multitask Learning: Heterogenous Patient Populations in the ICU

By Harini Suresh, Jen J. Gong, John V. Guttag. 2018

92 citations recorded by the source corpus.

Machine learning approaches have been effective in predicting adverse outcomes in different clinical settings. These models are often developed and evaluated on datasets with heterogeneous patient populations. However, good predictive performance on the aggregate population does not imply good performance for specific groups. In this work, we present a two-step framework to 1) learn relevant patient subgroups, and 2) predict an outcome for separate patient populations in a multi-task framework, where each population is a separate task. We demonstrate how to discover relevant groups in an unsupervised way with a sequence-to-sequence autoencoder. We show that using these groups in a multi-task framework leads to better predictive performance of in-hospital mortality both across groups and overall. We also highlight the need for more granular evaluation of performance when dealing with heterogeneous populations.

Subfields: Artificial Intelligence, Health Information Management, Meta & Continual Learning, Medical & Health AI

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Information Management

In Meta & Continual Learning

In Medical & Health AI

Citations in the corpus

FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age (2019, background): “In computer vision, it has been shown that popular large scale image datasets such as Imagenet are biased in terms of the origin of images (45% were from the U.S.) [56] or the underlying association between scene and race [52].”
A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle (2019, methodology): “Methods that adjust f include coupled learning methods, such as multitask learning, that parameterize different groups differently in the model definition and facilitate learning multiple functions that take into account group differences [16, 42].”
A Framework for Understanding Unintended Consequences of Machine Learning (2019, background): “…learning meth-ods , such as multitask learning, that parameterize different groups differently in the model deﬁnition and facilitate learning multiple simpler functions that take into account group differences (Gelman, 2006; Dwork et al., 2017; Suresh, Gong, and Guttag, 2018; Oneto et al., 2018).”
FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age for Bias Measurement and Mitigation (2021, background): “in Imagenet) [56] or biased underlying associations between scene and race in images [54].”
Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding (2021, background): “…phenotyping for discovering subgroups of patients with similar underlying disease mechanisms from temporal clinical data (Lasko et al., 2013; Suresh et al., 2018; Schulam et al., 2015), and disease progression modeling, for learning the hidden vector of comorbidities representing a disease…”
Empirical Analysis of Multi-Task Learning for Reducing Model Bias in Toxic Comment Detection (2019, methodology): “In terms of network architecture, our work is closest to the LSTM-based multi-task learning frameworks (Liang and Shu 2017; Suresh, Gong, and Guttag 2018).”