The Sums of Products Research GroupNOW HIRING a postdoc with excellent analytic skills, good programming skills, and publications at top venues of AI, ML, or algorithms, starting in 2019. Would you like to design the next generation algorithms for approximate counting, with applications to AI and ML? Helsinki has been ranked the 2nd safest city in the world (Mercer 2016) and the 6th most satisfied city in Europe (The Guardian 2014). If you got interested, send email to the group leader.
Click the image to view slides that give an idea of the kind of work done in the group!
Non-standard methodsespecially ones involving moderately exponential algorithmsfor evaluating large and structured sums of products have great prospects to significantly advance the state of the art in algorithm theory and computational statistics.
The group's current mission is to implement the vision by studying
- algorithm theory of computing sums of products,
- sums of products in computational statistics,
- applications in science and technology.
- Fast sum-product algorithms for set functions
- Efficient algorithms for structure learning in graphical models
- Bounding by products: global bounds via local views
- Mikko Koivisto, group leader
- Topi Talvitie, doctoral student (co-advisor Valentin Polishchuk)
- Jussi Viinikka, research assistant
- Aleksis Vuoksenmaa, research assistant
- Ralf Eggeling, post-doc (2015 - 2018)
- Sami Salonen, research assistant (summer 2017)
- Kustaa Kangas, doctoral student, post-doc (PhD 2016, co-advisor Matti Järvisalo)
- Teppo Niinimäki, doctoral student, post-doc (PhD 2015)
- Janne Korhonen, doctoral student, post-doc (PhD 2014, co-advisor Petteri Kaski)
- Esther Galbrun, doctoral student (PhD 2014, co-advisor Hannu TT Toivonen)
- Pekka Parviainen, doctoral student, post-doc (PhD 2012)
The recent research is funded mainly by the Academy of Finland (SEA 9/2014-8/2018; ATAC 9/2018-8/2022) and the University of Helsinki (grant of T.T., 1/2016-12/2019).
Last modified Oct 24, 2018.