Summary

My PhD research is focused on analyzing the dynamics of extreme events and their consequences on the biosphere. For this task I am using a particular Machine Learning family of models, called Reservoir Computing. More broadly, my interests lie in the analysis on nonlinear systems through the lens of machine learning.

Publications

For a full and always up to date list please check my Scholar profile.

  • Learning Extreme Vegetation Response to Climate Forcing: A Comparison of Recurrent Neural Network Architectures arxiv Martinuzzi, F., Mahecha, M. D., Camps-Valls, G., Montero, D., Williams, T., & Mora, K. EGUsphere, 2023, 1-32. (2023)

  • Data Cubes for Earth System Research: Challenges Ahead arxiv David Montero Loaiza , Guido Kraemer, …, Francesco Martinuzzi, …, Miguel Mahecha EarthArxiv (2023)

  • A standardized catalogue of spectral indices to advance the use of remote sensing in Earth system research html Montero, D., Aybar, C., Mahecha, M.D., Martinuzzi, F., Söchting, M. and Wieneke, S. Scientific Data, 10(1) (2023)

  • ReservoirComputing.jl: An Efficient and Modular Library for Reservoir Computing Models arXiv pdf Martinuzzi, F., Rackauckas, C., AbdelRehim, A., Mahecha, M. D. and Mora, K. Journal of Machine Learning Research (2022)

  • Composable and Reusable Neural Surrogates to Predict System Response of Causal Model Components pdf Anantharaman, R., AbdelRehim, A., Martinuzzi, F., Yalburgi, S., Fischer, K., Hertz, G., de Vos, P., Laughman, C., Ma, Y., Shah, V., Edelman, A. and Rackauckas, C. AAAI 2022 Workshop on AI for Design and Manufacturing (ADAM), (2022)

  • Composing Modeling and Simulation with Machine Learning in Julia arXiv Rackauckas, C., Anantharaman, R., Edelman, A., Gowda, S., Gwozdz, M., Jain, A., Laughman, C., Ma, Y., Martinuzzi, F., Pal, A., Rajput, U., Saba, E. and Shah, V. B. 14th Modelica Conference 2021, (2021)

Talks

  • Impact Predictability: Exploring Extremes in Biosphere Dynamics with Recurrent Neural Networks EGU24, Vienna, Austria

  • On Recurrent Neural Networks and Extreme Events: Application to Biosphere Dynamics 3rd ELLIS Workshop Machine Learning for Earth and Climate Sciences program, Valencia, Spain

  • An Introduction to Julia for Scientific Computing code slides Invited talk Parc Científic de la Universitat de València (2023)

  • Learning Biosphere Responses to Climate Drivers Using Echo State Observers slides Invited talk at the ELLIS Unit Jena Kickoff 2023

  • Chaotic time series predictions with ReservoirComputing.jl. Video JuliaCon 2021, (2021)

Notable Software

  • ReservoirComputing.jl: Julia implementation of a vast range of different Reservoir Computing algorithms.
  • CellularAutomata.jl: lightweight Julia implementation of Cellular Automata, both one dimensional and two dimensional.

Conferences and Summer Schools