Institution profile

University of Helsinki

Academic institutioneurope · fi
Official website
Research library357linked papers
Opportunities0open roles
Selected work

Representative Papers

Mixture of Experts Softens the Curse of Dimensionality in Operator Learning

Apr 13, 2024

To address the computational and memory bottlenecks imposed by the curse of dimensionality in high-dimensional operator learning, this paper proposes the Mixture-of-Experts Neural Operator (MoNO): a framework that decomposes a global nonlinear operator into multiple lightweight expert sub-operators, routed input-adaptively via a learnable decision-tree mechanism. Theoretically, we establish the first distributed universal approximation theorem, proving that MoNO uniformly approximates any Lipschitz-continuous nonlinear operator in Sobolev spaces; each expert’s depth, width, and rank scale as O(ε⁻¹), ensuring controllable memory footprint compatible with standard hardware. We further derive the first quantitative approximation rate for classical neural operators. Experiments and theory jointly demonstrate that MoNO achieves ε-accuracy with significantly reduced complexity, overcoming both expressive and deployability limitations inherent to monolithic neural operators.

20 citationsRead paper

Amortized Bayesian Workflow

Sep 06, 2024

Bayesian inference often faces a trade-off between computational efficiency and posterior accuracy, especially across multiple datasets. This paper proposes an adaptive hybrid inference workflow that—uniquely—integrates amortized variational inference (AVI) with Markov chain Monte Carlo (MCMC) in a dynamically coordinated manner. Leveraging principled posterior diagnostics, it constructs a Pareto frontier to enable automatic, optimal switching between AVI and MCMC. Computational reuse and scheduling optimization further boost inference throughput. The method unifies generative neural network modeling, MCMC refinement, and verifiable diagnostic mechanisms. Evaluated on tens of thousands of real and synthetic datasets, it achieves a 3.2× average speedup over standalone AVI or MCMC baselines, while preserving posterior fidelity—reducing KL divergence by 47% and increasing effective sample size (ESS) by 2.8×. This work delivers a scalable, efficient, and trustworthy solution for large-scale Bayesian inference.

2 citationsRead paper

Impact of Dataset Properties on Membership Inference Vulnerability of Deep Transfer Learning

Feb 07, 2024

This work investigates how dataset characteristics affect vulnerability to membership inference attacks (MIAs) in deep transfer learning, particularly for non-differentially private (non-DP) fine-tuned models. Method: We systematically quantify the impact of dataset size and per-class sample count on MIA success rates through empirical evaluation and theoretical modeling based on a simplified fine-tuning process. Contribution/Results: We establish, for the first time, that MIA advantage decays as a power law with respect to per-class sample count under fixed false positive rate—a finding that reveals the impractically large sample requirements needed to robustly protect the most vulnerable samples. This bridges a critical gap between DP-based privacy theory and real-world MIA threat models. Our results yield actionable, quantifiable guidelines for dataset-scale design in transfer learning, significantly enhancing the interpretability and controllability of privacy risks in non-DP settings.

2 citationsRead paper

Social, Spatial, and Self-Presence as Predictors of Basic Psychological Need Satisfaction in Social Virtual Reality

Feb 13, 2026

Extensive research has examined presence and basic psychological needs (drawing on Self-Determination Theory) in digital media. While prior work offers hints of potential connections, we lack a systematic account of whether and how distinct presence dimensions map onto the basic needs of autonomy, competence, and relatedness. We surveyed 301 social VR users and analyzed using Structural Equation Modeling. Results show that social presence predicts all three needs, while self-presence predicts competence and relatedness, and spatial presence shows no direct or moderating effects. Gender and age moderated these relationships: women benefited more from social presence for autonomy and relatedness, men from self- and spatial presence for competence and autonomy, and younger users showed stronger associations between social presence and relatedness, and between self-presence and autonomy. These findings position presence as a motivational mechanism shaped by demographic factors. The results offer theoretical insights and practical implications for designing inclusive, need-supportive multiuser VR environments.

1 citationsRead paper

Nonlinear Dynamic Factor Analysis With a Transformer Network

Jan 17, 2026

This study addresses the limitations of traditional linear dynamic factor models under nonlinear, non-Gaussian, and small-sample conditions. It proposes a novel nonlinear dynamic factor model by integrating the Transformer architecture into dynamic factor analysis. To enhance estimation stability in small samples, the approach incorporates a conventional factor model as a prior regularizer. The model leverages attention mechanisms to capture the time-varying contributions of individual variables and their lags to latent factors, thereby enabling the identification of economic regime shifts. Empirical results demonstrate that the proposed framework substantially outperforms standard methods in settings that deviate from linearity and Gaussianity, and it successfully constructs a coincident index of U.S. real economic activity.

1 citationsRead paper
Recent publications

Latest Papers