Multiple Instance Learning by Discriminative Training of Markov Networks

Hossein Hajimirsadeghi, jinling Li, Greg Mori, Tarek Sayed, Mohammad Zaki
Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, PMLR R11:412-421, 2013.

Abstract

We introduce a graphical framework for multiple instance learning (MIL) based on Markov networks. This framework can be used to model the traditional MIL definition as well as more general MIL definitions. Dif- ferent levels of ambiguity – the portion of positive instances in a bag – can be explored in weakly supervised data. To train these models, we propose a discriminative max- margin learning algorithm leveraging efficient inference for cardinality-based cliques. The efficacy of the proposed framework is evalu- ated on a variety of data sets. Experimental results verify that encoding or learning the degree of ambiguity can improve classifica- tion performance.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR11-hajimirsadeghi13a, title = {Multiple Instance Learning by Discriminative Training of {M}arkov Networks}, author = {Hajimirsadeghi, Hossein and Li, jinling and Mori, Greg and Sayed, Tarek and Zaki, Mohammad}, booktitle = {Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence}, pages = {412--421}, year = {2013}, editor = {Nicholson, Ann and Smyth, Padhraic}, volume = {R11}, series = {Proceedings of Machine Learning Research}, month = {12--14 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r11/main/assets/hajimirsadeghi13a/hajimirsadeghi13a.pdf}, url = {https://proceedings.mlr.press/r11/hajimirsadeghi13a.html}, abstract = {We introduce a graphical framework for multiple instance learning (MIL) based on Markov networks. This framework can be used to model the traditional MIL definition as well as more general MIL definitions. Dif- ferent levels of ambiguity – the portion of positive instances in a bag – can be explored in weakly supervised data. To train these models, we propose a discriminative max- margin learning algorithm leveraging efficient inference for cardinality-based cliques. The efficacy of the proposed framework is evalu- ated on a variety of data sets. Experimental results verify that encoding or learning the degree of ambiguity can improve classifica- tion performance.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Multiple Instance Learning by Discriminative Training of Markov Networks %A Hossein Hajimirsadeghi %A jinling Li %A Greg Mori %A Tarek Sayed %A Mohammad Zaki %B Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2013 %E Ann Nicholson %E Padhraic Smyth %F pmlr-vR11-hajimirsadeghi13a %I PMLR %P 412--421 %U https://proceedings.mlr.press/r11/hajimirsadeghi13a.html %V R11 %X We introduce a graphical framework for multiple instance learning (MIL) based on Markov networks. This framework can be used to model the traditional MIL definition as well as more general MIL definitions. Dif- ferent levels of ambiguity – the portion of positive instances in a bag – can be explored in weakly supervised data. To train these models, we propose a discriminative max- margin learning algorithm leveraging efficient inference for cardinality-based cliques. The efficacy of the proposed framework is evalu- ated on a variety of data sets. Experimental results verify that encoding or learning the degree of ambiguity can improve classifica- tion performance. %Z Reissued by PMLR on 04 October 2026.
APA
Hajimirsadeghi, H., Li, j., Mori, G., Sayed, T. & Zaki, M.. (2013). Multiple Instance Learning by Discriminative Training of Markov Networks. Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R11:412-421 Available from https://proceedings.mlr.press/r11/hajimirsadeghi13a.html. Reissued by PMLR on 04 October 2026.

Related Material