Federica Milinanni

I am a PhD student in Applied and Computational Mathematics. In my research I analyse and implement Markov Chain Monte Carlo methods (MCMC) for Uncertainty Quantification (UQ) of parameters in neuron models.  My research is part of the Brain-IT project within the Swedish e-Science Research Center (SeRC). I am also part of the Science for Life Laboratory (SciLifeLab).

PUBLICATIONS & PREPRINTS

• Milinanni, F. (2024). Alternative representation of the large deviation rate function and hyperparameter tuning schemes for Metropolis-Hastings Markov Chains.
• Milinanni, F., & Nyquist, P. (2024).On the large deviation principle for Metropolis-Hastings Markov Chains: the Lyapunov function condition and examples.
• Milinanni, F., & Nyquist, P. (2024).A large deviation principle for the empirical measures of Metropolis–Hastings chains.  Stochastic Processes and their Applications, 104293.
• Kramer, A., Milinanni, F., Nyquist, P., Jauhiainen, A., Eriksson, O. (2023).
  UQSA -- An R-Package for Uncertainty Quantification and Sensitivity Analysis for Biochemical Reaction Network Models.
• Eriksson, O., Kramer, A., Milinanni, F., & Nyquist, P. (2021).
  Sensitivity Approximation by the Peano-Baker Series.

THESES

• Milinanni, F. (2023). Markov Chain Monte Carlo Methods and Applications in Neuroscience. Licentiate Thesis.

CONFERENCES & WORKSHOPS

• MCQMC 2024, 18/08/2024-23/08/2024, Waterloo, Ontario (CANADA)
  Talk:"A large deviation principle for Metropolis-Hastings sampling"

• Bernoulli-IMS 11th World Congress in Probability and Statistics, 12/08/2024-16/08/2024, Bochum (GERMANY)
  Talk: "A large deviation principle for Metropolis-Hastings sampling"

• Algorithms & Computationally Intensive Inference seminars, 07/05/2024, University of Warwick, Department of Statistics (UK)
  Research Seminar: "Large deviation principle for Metropolis-Hastings sampling"

• SIAM Conference on Uncertainty Quantification (UQ24), 27/02/2024-01/03/2024, Trieste (ITALY)
  Talk: "Inverse Uncertainty Quantification and Forward Uncertainty Propagation in Reaction Network Models"

• Third workshop on Monte Carlo methods in Warsaw, 13/12/2023-15/12/2023, Warsaw (POLAND)
  Talk: "Large deviation principle for Metropolis-Hastings based Markov chains"

• Math + Neuroscience: Strengthening the Interplay Between Theory and Mathematics (semester program at ICERM), 17/09/2023-05/11/2023, Providence (RI, USA)
  Talk: "Parameter estimation with uncertainty quantification via Markov Chain Monte Carlo methods"

• Licentiate Seminar, 14/09/2023, KTH,
  "Markov Chain Monte Carlo Methods and Applications in Neuroscience"

• Monte Carlo Methods and Applications, 26-30/06/2023, Paris (FRANCE)
  Poster: "Large Deviation Principle for the Metropolis-Hastings algorithm"
• EBRAINS RI Training: Tools for Molecular Simulation of Neuronal Signaling Cascades, 21-23/06/2023, Hybrid
  Talk: "Subcellular toolset"
• Human Brain Project Summit 2023, 28-31/03/2023, Marseille (FRANCE)
  Poster: "Data-driven signaling pathway modelling"
• Bayes Comp 2023, 15-17/03/2023, Levi (FINLAND)
  Talk:"Large Deviation Principle for the Metropolis-Hastings algorithm"
• Hausdorff School on Inverse Problems for Multi-Scale Models, 22-26/08/2022, Bonn (GERMANY)
  Talk: "Uncertainty Quantification in Neuronal Signalling Pathways"
• Monte Carlo and Quasi-Monte Carlo Methods in Scientific Computing, 17-22/07/2022, Linz (AUSTRIA)
• Stockholms Mathematikcentrum Pre-Colloquium, 25/05/2022, Albanova, Stockholm (SWEDEN).
  Talk: "Predicting graphs: Building blocks"
• Training Workshop on Tools for Molecular Simulation of Neuronal Signaling Cascades, 07-10/03/2022, Virtual
  Talk: "Subcellular model building toolset"
• YAMC2021: Conference Of Young Applied Mathematicians, 13-17/09/2021, Leuca, Lecce (ITALY)
  Talk: "Approximating Fisher Information for efficient MCMC"
  Poster: "Approximating Fisher Information for efficient MCMC"
• INCOME2021 (Integrative pathway modeling in systems biology and systems medicine), 01-04/03/2021, Virtual
  Poster: "Approximating Fisher Information for efficient MCMC"