KEPACO Software

The KEPACO group is part of Aalto from since beginning of 2012. The group develops machine learning methods, models and tools for computational sciences, in particular computational biology. The methodological backbone of the group is kernel methods and regularized learning. The group particularly focusses in learning with multiple and structured targets, multiple views and ensembles. Applications of interest in computational biology include network reconstruction, gene functional classification as well as biomarker discovery.

Metabolite identification through machine learning.

Screen shot of Fingerid server

SPIN: Structured Prediction of Network Response

Learning curves of different aggregation approaches.
  • Related publication:
    • Su, Hongyu; Gionis, Aristides; Rousu, Juho. Structured Prediction of Network Response. In Proceedings of the 31th International Conference on Machine Learning (ICML 2014), Journal of Machine Learning Research (JMLR) W&CP 32:442-450 [original] [code]
  • Package is located in Github, and can be cloned via
    • HTTP: https://github.com/hongyusu/SPIN.git
    • SSH: git@github.com:hongyusu/SPIN.git
    • SVN: https://github.com/hongyusu/SPIN

MAM: Structured Output Learning with Random Output Graph Aggregation

Learning curves of different aggregation approaches.
  • Related publication:
    • Su, Hongyu; Rousu, Juho. Multilabel Classification through Random Graph Ensembles. In Machine Learning, DOI: 10.1007/s10994-014-5465-9. [online][preprint]
    • Su, Hongyu; Rousu, Juho. Multilabel Classification through Random Graph Ensembles. In Proceedings of the 5th Asian Conference on Machine Learning (ACML 2013), Journal of Machine Learning Research (JMLR) W&CP 29:404-418 [original]
  • Package is located in Github, and can be cloned via
    • HTTP: https://github.com/hongyusu/RandomOutputGraphEnsemble.git
    • SSH: git@github.com:hongyusu/RandomOutputGraphEnsemble.git
    • SVN: https://github.com/hongyusu/RandomOutputGraphEnsemble

Back to group home page