Multi-prototype Support Vector Machine

Aiolli, Fabio; Sperduti, Alessandro

Multi-prototype Support Vector Machine

Download BibTeX Cite

Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence. Pages 541-545.

Abstract

We extend multiclass SVM to multiple prototypes per class. For this framework, we give a compact constrained quadratic problem and we suggest an efficient algorithm for its optimization that guarantees a local minimum of the objective function. An annealed process is also proposed that helps to escape from local minima. Finally, we report experiments where the performance obtained using linear models is almost comparable to that obtained by state-of-art kernel-based methods but with a significant reduction (of one or two orders) in response time.

Fabio Aiolli, and Alessandro Sperduti. 2003. Multi-prototype Support Vector Machine. In Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, IJCAI-03. IJCAI, 541–545. https://ijcai.org/proceedings/2003/80

@inproceedings{ijcai2003p80,
  title     = {Multi-prototype Support Vector Machine},
  author    = {Aiolli, Fabio and Sperduti, Alessandro},
  booktitle = {Proceedings of the Eighteenth International Joint Conference on
               Artificial Intelligence, {IJCAI-03}},
  publisher = {International Joint Conferences on Artificial Intelligence},
  editor    = {Georg Gottlob},
  pages     = {541--545},
  year      = {2003},
  url       = {https://ijcai.org/proceedings/2003/80},
}

TY  - CPAPER
TI  - Multi-prototype Support Vector Machine
AU  - Aiolli, Fabio
AU  - Sperduti, Alessandro
T2  - Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, IJCAI-03
PB  - International Joint Conferences on Artificial Intelligence
PY  - 2003
SP  - 541
EP  - 545
UR  - https://ijcai.org/proceedings/2003/80
ER  -