Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health
© 2010-2012 IEEE. While accurately predicting mood and wellbeing could have a number of important clinical benefits, traditional machine learning (ML) methods frequently yield low performance in this domain. We posit that this is because a one-size-fits-all machine learning model is inherently ill-s...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2021
|
| Online Access: | https://hdl.handle.net/1721.1/133994 |
| _version_ | 1826212811841208320 |
|---|---|
| author | Taylor, Sara Jaques, Natasha Nosakhare, Ehimwenma Sano, Akane Picard, Rosalind |
| author_facet | Taylor, Sara Jaques, Natasha Nosakhare, Ehimwenma Sano, Akane Picard, Rosalind |
| author_sort | Taylor, Sara |
| collection | MIT |
| description | © 2010-2012 IEEE. While accurately predicting mood and wellbeing could have a number of important clinical benefits, traditional machine learning (ML) methods frequently yield low performance in this domain. We posit that this is because a one-size-fits-all machine learning model is inherently ill-suited to predicting outcomes like mood and stress, which vary greatly due to individual differences. Therefore, we employ Multitask Learning (MTL) techniques to train personalized ML models which are customized to the needs of each individual, but still leverage data from across the population. Three formulations of MTL are compared: i) MTL deep neural networks, which share several hidden layers but have final layers unique to each task; ii) Multi-task Multi-Kernel learning, which feeds information across tasks through kernel weights on feature types; and iii) a Hierarchical Bayesian model in which tasks share a common Dirichlet Process prior. We offer the code for this work in open source. These techniques are investigated in the context of predicting future mood, stress, and health using data collected from surveys, wearable sensors, smartphone logs, and the weather. Empirical results demonstrate that using MTL to account for individual differences provides large performance improvements over traditional machine learning methods and provides personalized, actionable insights. |
| first_indexed | 2024-09-23T15:38:17Z |
| format | Article |
| id | mit-1721.1/133994 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T15:38:17Z |
| publishDate | 2021 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1339942022-04-01T16:26:07Z Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health Taylor, Sara Jaques, Natasha Nosakhare, Ehimwenma Sano, Akane Picard, Rosalind © 2010-2012 IEEE. While accurately predicting mood and wellbeing could have a number of important clinical benefits, traditional machine learning (ML) methods frequently yield low performance in this domain. We posit that this is because a one-size-fits-all machine learning model is inherently ill-suited to predicting outcomes like mood and stress, which vary greatly due to individual differences. Therefore, we employ Multitask Learning (MTL) techniques to train personalized ML models which are customized to the needs of each individual, but still leverage data from across the population. Three formulations of MTL are compared: i) MTL deep neural networks, which share several hidden layers but have final layers unique to each task; ii) Multi-task Multi-Kernel learning, which feeds information across tasks through kernel weights on feature types; and iii) a Hierarchical Bayesian model in which tasks share a common Dirichlet Process prior. We offer the code for this work in open source. These techniques are investigated in the context of predicting future mood, stress, and health using data collected from surveys, wearable sensors, smartphone logs, and the weather. Empirical results demonstrate that using MTL to account for individual differences provides large performance improvements over traditional machine learning methods and provides personalized, actionable insights. 2021-10-27T19:57:33Z 2021-10-27T19:57:33Z 2020 2019-07-31T16:49:32Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/133994 en 10.1109/TAFFC.2017.2784832 IEEE Transactions on Affective Computing Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT web domain |
| spellingShingle | Taylor, Sara Jaques, Natasha Nosakhare, Ehimwenma Sano, Akane Picard, Rosalind Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health |
| title | Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health |
| title_full | Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health |
| title_fullStr | Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health |
| title_full_unstemmed | Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health |
| title_short | Personalized Multitask Learning for Predicting Tomorrow's Mood, Stress, and Health |
| title_sort | personalized multitask learning for predicting tomorrow s mood stress and health |
| url | https://hdl.handle.net/1721.1/133994 |
| work_keys_str_mv | AT taylorsara personalizedmultitasklearningforpredictingtomorrowsmoodstressandhealth AT jaquesnatasha personalizedmultitasklearningforpredictingtomorrowsmoodstressandhealth AT nosakhareehimwenma personalizedmultitasklearningforpredictingtomorrowsmoodstressandhealth AT sanoakane personalizedmultitasklearningforpredictingtomorrowsmoodstressandhealth AT picardrosalind personalizedmultitasklearningforpredictingtomorrowsmoodstressandhealth |