SHiNe
Statistical Theory for High-Dimensional Structured Network Models
SHiNe develops statistical theory for structured network data. The project studies community recovery in networks with temporal, spatial, high-dimensional, and higher-order structure, focusing on the fundamental limits of statistical and computationally efficient inference.
Research themes
- Structured network models
- Community recovery
- Temporal and higher-order networks
- Node attributes and geometric representations
- Information-theoretic and computational limits
Scientific questions
- How do temporal dynamics, spatial structure, and node covariates affect the recoverability of latent communities?
- Under what conditions is consistent cluster recovery possible in structured network models?
- What information-theoretic and computational barriers govern network inference?
Team
- Lasse Leskelä, principal investigator
- Ian Välimaa, doctoral researcher (2024–)
- Vilma Moilanen, doctoral researcher (starting September 2026)
- NN, postdoctoral researcher (position planned for 2027)
Open positions
There are currently no open positions.
The project plans to recruit a postdoctoral researcher in 2027. Prospective candidates with interests in network inference, stochastic processes, high-dimensional statistics, or related areas are encouraged to contact the PI informally at lasse.leskela@aalto.fi.
Publications
Publications and preprints related to the project will be listed here.
Funding
This project is funded by the Research Council of Finland and Aalto University.
