Marginal Structured SVM with Hidden Variables

Wei Ping, Qiang Liu, Alex Ihler
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):190-198, 2014.

Abstract

In this work, we propose the marginal structured SVM (MSSVM) for structured prediction with hidden variables. MSSVM properly accounts for the uncertainty of hidden variables, and can significantly outperform the previously proposed latent structured SVM (LSSVM; Yu & Joachims (2009)) and other state-of-art methods, especially when that uncertainty is large. Our method also results in a smoother objective function, making gradient-based optimization of MSSVMs converge significantly faster than for LSSVMs. We also show that our method consistently outperforms hidden conditional random fields (HCRFs; Quattoni et al. (2007)) on both simulated and real-world datasets. Furthermore, we propose a unified framework that includes both our and several other existing methods as special cases, and provides insights into the comparison of different models in practice.

Cite this Paper


BibTeX
@InProceedings{pmlr-v32-ping14, title = {Marginal Structured SVM with Hidden Variables}, author = {Ping, Wei and Liu, Qiang and Ihler, Alex}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {190--198}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/ping14.pdf}, url = {https://proceedings.mlr.press/v32/ping14.html}, abstract = {In this work, we propose the marginal structured SVM (MSSVM) for structured prediction with hidden variables. MSSVM properly accounts for the uncertainty of hidden variables, and can significantly outperform the previously proposed latent structured SVM (LSSVM; Yu & Joachims (2009)) and other state-of-art methods, especially when that uncertainty is large. Our method also results in a smoother objective function, making gradient-based optimization of MSSVMs converge significantly faster than for LSSVMs. We also show that our method consistently outperforms hidden conditional random fields (HCRFs; Quattoni et al. (2007)) on both simulated and real-world datasets. Furthermore, we propose a unified framework that includes both our and several other existing methods as special cases, and provides insights into the comparison of different models in practice.} }
Endnote
%0 Conference Paper %T Marginal Structured SVM with Hidden Variables %A Wei Ping %A Qiang Liu %A Alex Ihler %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-ping14 %I PMLR %P 190--198 %U https://proceedings.mlr.press/v32/ping14.html %V 32 %N 2 %X In this work, we propose the marginal structured SVM (MSSVM) for structured prediction with hidden variables. MSSVM properly accounts for the uncertainty of hidden variables, and can significantly outperform the previously proposed latent structured SVM (LSSVM; Yu & Joachims (2009)) and other state-of-art methods, especially when that uncertainty is large. Our method also results in a smoother objective function, making gradient-based optimization of MSSVMs converge significantly faster than for LSSVMs. We also show that our method consistently outperforms hidden conditional random fields (HCRFs; Quattoni et al. (2007)) on both simulated and real-world datasets. Furthermore, we propose a unified framework that includes both our and several other existing methods as special cases, and provides insights into the comparison of different models in practice.
RIS
TY - CPAPER TI - Marginal Structured SVM with Hidden Variables AU - Wei Ping AU - Qiang Liu AU - Alex Ihler BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-ping14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 190 EP - 198 L1 - http://proceedings.mlr.press/v32/ping14.pdf UR - https://proceedings.mlr.press/v32/ping14.html AB - In this work, we propose the marginal structured SVM (MSSVM) for structured prediction with hidden variables. MSSVM properly accounts for the uncertainty of hidden variables, and can significantly outperform the previously proposed latent structured SVM (LSSVM; Yu & Joachims (2009)) and other state-of-art methods, especially when that uncertainty is large. Our method also results in a smoother objective function, making gradient-based optimization of MSSVMs converge significantly faster than for LSSVMs. We also show that our method consistently outperforms hidden conditional random fields (HCRFs; Quattoni et al. (2007)) on both simulated and real-world datasets. Furthermore, we propose a unified framework that includes both our and several other existing methods as special cases, and provides insights into the comparison of different models in practice. ER -
APA
Ping, W., Liu, Q. & Ihler, A.. (2014). Marginal Structured SVM with Hidden Variables. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):190-198 Available from https://proceedings.mlr.press/v32/ping14.html.

Related Material