Riikka Huusari

Post-doctoral researcher (Aalto University, Finland)

I am focusing my research to kernel methods, considering both the traditional scalar-valued kernels and the more general operator-valued kernels.

Preprints:

Riikka Huusari, Cécile Capponi, Paul Villoutreix and Hachem Kadri: Kernel transfer over multiple views for missing data completion October 2019. (PDF, Python code)

Book chapters and journal papers:

Riikka Huusari, Hachem Kadri and Cécile Capponi: General Framework for Multi-view Metric Learning in Linking and Mining Heterogeneous and Multi-view Data, chapter 11, p.265-294, Springer, January 2019

Publications in international conferences:

Hachem Kadri, Stéphane Ayache, Riikka Huusari, Alain Rakotomamonjy and Liva Ralaivola: Partial Trace Regression and Low-Rank Kraus Decomposition in 37th International Conference on Machine Learning (ICML), Vienna, Austria, July 2020. (PDF, Python code)

Riikka Huusari and Hachem Kadri: Entangled Kernels in 28th International Joint Conference on Artificial Intelligence (IJCAI), Macao, China, August 2019. (PDF, BibTeX, Python code, poster, presentation)

Riikka Huusari, Hachem Kadri and Cécile Capponi: Multi-View Metric Learning in Vector-Valued Kernel Spaces in The 21st International Conference on Artificial Intelligence and Statistics (AISTATS), Lanzarote, Spain, April 2018. (PDF, BibTeX, Python code, Poster)

Publications in French conferences:

Riikka Huusari, Cécile Capponi, Hachem Kadri and Paul Villoutreix: Apprentissage multi-vues pour la complétion transmodale de matrices de noyaux in Conférence sur l'Apprentissage automatique (CAp), Toulouse, France, July 2019

Riikka Huusari and Hachem Kadri: Intrication et noyaux à valeurs opérateurs in Conférence sur l'Apprentissage automatique (CAp), Rouen, France, June 2018

Riikka Huusari, Hachem Kadri and Cécile Capponi: Support Vector Machine Framework for Multi-View Metric Learning. in 50e Journées de Statistique de la Société Française de Statistique, Paris, France, May-June 2018

Riikka Huusari, Hachem Kadri and Cécile Capponi: Noyaux à valeurs opérateurs et apprentissage de métriques multi-vues. in Conférence sur l'Apprentissage automatique (CAp), Grenoble, France, June 2017

Workshop publications:

Riikka Huusari, Cécile Capponi, Hachem Kadri and Paul Villoutreix: Cross-view kernel transfer in Data and Machine Learning Advances with Multiple Views (DAMVL) Workshop of of ECMLPKDD 2019, Würzburg, Germany, September 2019

Thesis:

Kernel learning for structured data: A study on learning operator- and scalar-valued kernels for multi-view and multi-task learning problems November 2019. (link)