Juha Harviainen
I’m a postdoctoral researcher working at the University of Helsinki in the Sums of Products research group, where I also completed my doctoral degree in 2024 under the supervision of professor Mikko Koivisto. My research focuses currently on parameterized complexity and perfect sampling, and I’m interested in randomized algorithms, complexity theory and information theory more generally.
Research
- Identifying All Snarls and Superbubbles in Linear-Time, via a Unified SPQR-tree Framework. Francisco Sena, Aleksandr Politov, Corentin Moumard, Manuel Cáceres, Sebastian Schmidt, Juha Harviainen, Alexandru I. Tomescu. 2025. arXiv preprint.
- Scaling Up Bayesian DAG Sampling. Daniele Nikzad, Alexander Zhilkin, Juha Harviainen, Jack Kuipers, Giusi Moffa, Mikko Koivisto. 2025. arXiv preprint.
- Improving Decision Trees through the Lens of Parameterized Local Search. Juha Harviainen, Frank Sommer, Manuel Sorge. NeurIPS 2025 (to appear).
- Graph Reconstruction with the Connected Components Oracle. Juha Harviainen, Pekka Parviainen. 2025. arXiv preprint.
- Quantum Speedups for Bayesian Network Structure Learning. Juha Harviainen, Kseniya Rychkova, Mikko Koivisto. UAI 2025.
- Optimal Decision Tree Pruning Revisited: Algorithms and Complexity. Juha Harviainen, Frank Sommer, Manuel Sorge, Stefan Szeider. ICML 2025.
- On Tractability of Learning Bayesian Networks with Ancestral Constraints. Juha Harviainen, Pekka Parviainen. AISTATS 2025.
- Estimating the Permanent by Nesting Importance Sampling. Juha Harviainen, Mikko Koivisto. ICML 2024.
- Faster Perfect Sampling of Bayesian Network Structures. Juha Harviainen, Mikko Koivisto. UAI 2024.
- Revisiting Bayesian Network Learning with Small Vertex Cover. Juha Harviainen, Mikko Koivisto. UAI 2023.
- On Inference and Learning With Probabilistic Generating Circuits. Juha Harviainen, Vaidyanathan Peruvemba Ramaswamy, Mikko Koivisto. UAI 2023.
- A Faster Practical Approximation Scheme for the Permanent. Juha Harviainen, Mikko Koivisto. AAAI 2023.
- Trustworthy Monte Carlo. Juha Harviainen, Mikko Koivisto, Petteri Kaski. NeurIPS 2022.
- Approximating the Permanent with Deep Rejection Sampling. Juha Harviainen, Antti Röyskö, Mikko Koivisto. NeurIPS 2021.
- Software Framework for Data Fault Injection to Test Machine Learning Systems. Jukka K. Nurminen, Tuomas Halvari, Juha Harviainen, Juha Mylläri, Antti Röyskö, Juuso Silvennoinen, Tommi Mikkonen. ISSRE Workshops 2019.
Other
- Tie koodariksi. A website for Finnish schools for teaching programming made by me, Antti Laaksonen, Roope Salmi and Topi Talvitie.
- Advances in Sampling and Counting Bipartite Matchings and Directed Acyclic Graphs. Juha Harviainen. 2024. Doctoral dissertation.
Contact
You can contact me via email:
juha.harviainen@helsinki.fi
Juha Harviainen 