dc.contributor.author | Wang, Shenhao | |
dc.contributor.author | Mo, Baichuan | |
dc.contributor.author | Zhao, Jinhua | |
dc.date.accessioned | 2022-02-07T20:48:33Z | |
dc.date.available | 2022-02-07T19:10:21Z | |
dc.date.available | 2022-02-07T20:48:33Z | |
dc.date.issued | 2021-04 | |
dc.date.submitted | 2021-02 | |
dc.identifier.issn | 0191-2615 | |
dc.identifier.uri | https://hdl.handle.net/1721.1/140214.2 | |
dc.description.abstract | Researchers often treat data-driven and theory-driven models as two disparate or even conflicting methods in travel behavior analysis. However, the two methods are highly complementary because data-driven methods are more predictive but less interpretable and robust, while theory-driven methods are more interpretable and robust but less predictive. Using their complementary nature, this study designs a theory-based residual neural network (TB-ResNet) framework, which synergizes discrete choice models (DCMs) and deep neural networks (DNNs) based on their shared utility interpretation. The TB-ResNet framework is simple, as it uses a ( 1-) weighting to take advantage of DCMs’ simplicity and DNNs’ richness, and to prevent underfitting from the DCMs and overfitting from the DNNs. This framework is also flexible: three instances of TB-ResNets are designed based on multinomial logit model (MNL-ResNets), prospect theory (PT-ResNets), and hyperbolic discounting (HD-ResNets), which are tested on three data sets. Compared to pure DCMs, the TB-ResNets provide greater prediction accuracy and reveal a richer set of behavioral mechanisms owing to the utility function augmented by the DNN component in the TB-ResNets. Compared to pure DNNs, the TB-ResNets can modestly improve prediction and significantly improve interpretation and robustness, because the DCM component in the TB-ResNets stabilizes the utility functions and input gradients. Overall, this study demonstrates that it is both feasible and desirable to synergize DCMs and DNNs by combining their utility specifications under a TB-ResNet framework. Although some limitations remain, this TB-ResNet framework is an important first step to create mutual benefits between DCMs and DNNs for travel behavior modeling, with joint improvement in prediction, interpretation, and robustness. | en_US |
dc.language.iso | en | |
dc.publisher | Elsevier BV | en_US |
dc.relation.isversionof | http://dx.doi.org/10.1016/j.trb.2021.03.002 | en_US |
dc.rights | Creative Commons Attribution-NonCommercial-NoDerivs License | en_US |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/ | en_US |
dc.source | MIT web domain | en_US |
dc.title | Theory-based residual neural networks: A synergy of discrete choice models and deep neural networks | en_US |
dc.type | Article | en_US |
dc.identifier.citation | Wang, Shenhao, Mo, Baichuan and Zhao, Jinhua. 2021. "Theory-based residual neural networks: A synergy of discrete choice models and deep neural networks." Transportation Research Part B: Methodological, 146. | en_US |
dc.contributor.department | Massachusetts Institute of Technology. Department of Urban Studies and Planning | |
dc.relation.journal | Transportation Research Part B: Methodological | en_US |
dc.eprint.version | Original manuscript | en_US |
dc.type.uri | http://purl.org/eprint/type/JournalArticle | en_US |
eprint.status | http://purl.org/eprint/status/NonPeerReviewed | en_US |
dc.date.updated | 2022-02-07T18:57:02Z | |
dspace.orderedauthors | Wang, S; Mo, B; Zhao, J | en_US |
dspace.date.submission | 2022-02-07T18:57:04Z | |
mit.journal.volume | 146 | en_US |
mit.license | PUBLISHER_CC | |
mit.metadata.status | Authority Work Needed | en_US |