Aapo Hyvärinen

Menu

New:

Arxiv: Multi-View Causal Discovery Without Non-Gaussianity: Identifiability and Algorithms

TMLR: A noise-corrected Langevin algorithm and sampling by half-denoising

Imag Neurosci: Second-order instantaneous causal analysis of spontaneous MEG

ICML2025: Density Ratio Estimation with Conditional Probability Paths

Home page

Professor of Computer Science (Machine Learning)

Helsinki Probabilistic Machine Learning Lab

Books

• Painful intelligence: What AI can tell us about human suffering (2022/24)
• Natural Image Statistics (2009)
• Independent Component Analysis (2001)

Recent reviews (2023)

• Nonlinear independent component analysis for principled disentanglement
• Identifiability of latent-variable and structural-equation models

A randomized research manifesto ?