Learning Tasks for Multitask Learning
© 2018 Copyright held by the owner/author(s). 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...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
ACM
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137636 |
| _version_ | 1826188004137369600 |
|---|---|
| author | Suresh, Harini Gong, Jen J. Guttag, John V. |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Suresh, Harini Gong, Jen J. Guttag, John V. |
| author_sort | Suresh, Harini |
| collection | MIT |
| description | © 2018 Copyright held by the owner/author(s). 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. |
| first_indexed | 2024-09-23T07:53:07Z |
| format | Article |
| id | mit-1721.1/137636 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T07:53:07Z |
| publishDate | 2021 |
| publisher | ACM |
| record_format | dspace |
| spelling | mit-1721.1/1376362023-01-30T20:55:18Z Learning Tasks for Multitask Learning Heterogenous Patient Populations in the ICU Suresh, Harini Gong, Jen J. Guttag, John V. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © 2018 Copyright held by the owner/author(s). 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. 2021-11-08T12:50:31Z 2021-11-08T12:50:31Z 2018-07-19 2019-05-30T14:35:44Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137636 Suresh, Harini, Gong, Jen J. and Guttag, John V. 2018. "Learning Tasks for Multitask Learning." en 10.1145/3219819.3219930 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf ACM arXiv |
| spellingShingle | Suresh, Harini Gong, Jen J. Guttag, John V. Learning Tasks for Multitask Learning |
| title | Learning Tasks for Multitask Learning |
| title_full | Learning Tasks for Multitask Learning |
| title_fullStr | Learning Tasks for Multitask Learning |
| title_full_unstemmed | Learning Tasks for Multitask Learning |
| title_short | Learning Tasks for Multitask Learning |
| title_sort | learning tasks for multitask learning |
| url | https://hdl.handle.net/1721.1/137636 |
| work_keys_str_mv | AT sureshharini learningtasksformultitasklearning AT gongjenj learningtasksformultitasklearning AT guttagjohnv learningtasksformultitasklearning AT sureshharini heterogenouspatientpopulationsintheicu AT gongjenj heterogenouspatientpopulationsintheicu AT guttagjohnv heterogenouspatientpopulationsintheicu |