Tuomo Alasalmi

Tuomo Alasalmi

MSc. (Tech)

Doctoral researcher

Biography

I am a doctoral researcher in machine learning at the Biomimetics and Intelligent Systems Group (BISG). I am especially interested in uncertainty of classifier predictions and my doctoral thesis, which is now nearly finished, deals with that uncertainty on limited (small or incomplete) data sets. In addition to the more theoretical work, I have also applied my knowledge e.g. in modelling eye disease risk based on routine eye checkup data in a project called CRYSTAL which was funded by Business Finland.

Research interests

  • Classifier uncertainty and calibration
  • Machine learning in health applications

Teaching

  • Towards Data Mining (521156S): 2017-2020
  • Operating systems (521453A): 2013-2019

Research groups

  • Biomimetics and Intelligent Systems

Selected publications

  • Alasalmi, Tuomo; Suutala, Jaakko; Röning, Juha; Koskimäki, Heli (2020) Better classifier calibration for small datasets. - ACM transactions on knowledge discovery from data 14 (3), 34:1-34:19 . [Original]
  • Alasalmi, Tuomo; Koskimäki, Heli; Suutala, Jaakko; Röning, Juha (2018) Getting More Out of Small Data Sets : Improving the Calibration Performance of Isotonic Regression by Generating More Data. (Artikkeli tieteellisessä konferenssijulkaisussa). - Proceedings of the 10th International Conference on Agents and Artificial Intelligence : Volume 2 : ICAART. Ei sarjaa/No series. 379-386. [Original] [Self-archived]
  • Alasalmi, Tuomo; Koskimäki, Heli; Suutala, Jaakko; Röning, Juha (2016) Instance Level Classification Confidence Estimation. (Artikkeli tieteellisessä konferenssijulkaisussa). - Distributed Computing and Artificial Intelligence, 13th International Conference : Part I. Advances in Intelligent Systems and Computing 474. 275-282. [Original]
  • Alasalmi, Tuomo; Koskimäki, Heli; Suutala, Jaakko; Röning, Juha (2015) Classification uncertainty of multiple imputed data. (Artikkeli tieteellisessä konferenssijulkaisussa). - 2015 IEEE Symposium Series on Computational Intelligence. SSCI 2015. 8-10 December 2015 Cape Town, South Africa. Ei sarjaa/No series. 151-158. [Original]