My Google Scholar profile.
2013 |
Dembczynski, Krzysztof; Jachnik, Arkadiusz; Kotlowski, Wojciech; Waegeman, Willem; Huellermeier, Eyke (2013): Optimizing the F-Measure in Multi-Label Classification: Plug-in Rule Approach versus Structured Loss Minimization. In: Proceedings of the International Conference on Machine Learning, 2013. (Type: Inproceeding | Abstract | Links | BibTeX)@inproceedings{Dembczynski_et_al_2013,
title = {Optimizing the F-Measure in Multi-Label Classification: Plug-in Rule Approach versus Structured Loss Minimization},
author = {Krzysztof Dembczynski and Arkadiusz Jachnik and Wojciech Kotlowski and Willem Waegeman and Eyke Huellermeier},
url = {http://jmlr.org/proceedings/papers/v28/dembczynski13.pdf},
year = {2013},
date = {2013-06-19},
booktitle = {Proceedings of the International Conference on Machine Learning},
abstract = {We compare the plug-in rule approach for optimizing the F-measure in multi-label classification with an approach based on structured loss minimization, such as the structured support vector machine (SSVM). Whereas the former derives an optimal prediction from a probabilistic model in a separate inference step, the latter seeks to optimize the F-measure directly during the training phase. We introduce a novel plug-in rule algorithm that estimates all parameters required for a Bayes-optimal prediction via a set of multinomial regression models, and we compare this algorithm with SSVMs in terms of computational complexity and statistical consistency. As a main theoretical result, we show that our plug-in rule algorithm is consistent, whereas the SSVM approaches are not. Finally, we present results of a large experimental study showing the benefits of the introduced algorithm.},
keywords = {}
}
We compare the plug-in rule approach for optimizing the F-measure in multi-label classification with an approach based on structured loss minimization, such as the structured support vector machine (SSVM). Whereas the former derives an optimal prediction from a probabilistic model in a separate inference step, the latter seeks to optimize the F-measure directly during the training phase. We introduce a novel plug-in rule algorithm that estimates all parameters required for a Bayes-optimal prediction via a set of multinomial regression models, and we compare this algorithm with SSVMs in terms of computational complexity and statistical consistency. As a main theoretical result, we show that our plug-in rule algorithm is consistent, whereas the SSVM approaches are not. Finally, we present results of a large experimental study showing the benefits of the introduced algorithm.
|
2012 |
Jachnik, Arkadiusz; Szwabe, Andrzej; Misiorek, Pawel; Walkowiak, Przemyslaw (2012): TOAST Results for OAEI 2012. In: Ontology Matching, pp. 205, 2012. (Type: Article | BibTeX)@article{jachniktoast,
title = {TOAST Results for OAEI 2012},
author = {Jachnik, Arkadiusz and Szwabe, Andrzej and Misiorek, Pawel and Walkowiak, Przemyslaw},
year = {2012},
date = {2012-11-12},
journal = {Ontology Matching},
pages = {205},
keywords = {}
}
|
2011 |
Szwabe, Andrzej; Jachnik, Arkadiusz; Figaj, Andrzej; Blinkiewicz, Michal (2011): Semantic Structure Matching Recommendation Algorithm. In: pp. 73–81, 2011. (Type: Article | BibTeX)@article{szwabe2011semantic,
title = {Semantic Structure Matching Recommendation Algorithm},
author = {Szwabe, Andrzej and Jachnik, Arkadiusz and Figaj, Andrzej and Blinkiewicz, Michal},
year = {2011},
date = {2011-01-01},
booktitle = {Multimedia Communications, Services and Security},
pages = {73--81},
publisher = {Springer},
keywords = {}
}
|
Video of my talks at conferences:
Optimizing the F-measure in Multi-label Classification: Plug-in Rule Approach versus Structured Loss Minimization
International Conference on Machine Learning 2013 (Atlanta, USA)
techtalks.tv