Francisco Richter Mendoza

Research

Evolutionary biology

Statistical: fitting models of speciation and extinction to trees of living species by simulating the lineages the tree is missing, and working out what such trees can and cannot tell us.

Biological: whether diversity limits itself, whether a lineage's age or distinctiveness changes its fate, and whether diversification rates are inherited.

Robust and causal regularization

With Ernst Wit: which relationships still hold when conditions change, in data where the conditions were not recorded. Applied to learning outcomes in Peru's national assessment (Conference on Complex Systems, 2026) and to problems in social and network science.

Other systems

Phylogenies of stellar chemical abundances, microbiome interaction networks, and injury mortality in Chile.

Publications

1. 2026
   Where the tree of life is empirically resolved, and where it is not: an open atlas of species-level phylogenies and their archival uncertainty F. Richter bioRxiv
2. 2025
   Neural causal regularization: extending causal invariance to deep models F. Richter, K. Rigana, E. C. Wit Statistics for Innovation I, 97–102. Springer Nature
3. 2024
   Random graphical model of microbiome interactions in related environments V. Vinciotti, E. C. Wit, F. Richter Journal of Agricultural, Biological and Environmental Statistics
4. 2021
   Detecting phylodiversity-dependent diversification with a general phylogenetic inference framework F. Richter, T. Janzen, H. Hildenbrandt, E. C. Wit, R. S. Etienne bioRxiv
5. 2021
   Microbiome and environment explain the absence of correlations between consumers and their diet in Bornean microsnails K. P. Hendriks, …, F. J. Richter Mendoza, M. Schilthuizen, R. S. Etienne Ecology 102(2), e03237
6. 2021
   Plant diets of land snail community members are similar in composition but differ in richness K. P. Hendriks, …, F. J. Richter Mendoza, M. Schilthuizen, R. S. Etienne Journal of Molluscan Studies 87(4), eyab041
7. 2020
   Introducing a general class of species diversification models for phylogenetic trees F. Richter, B. Haegeman, R. S. Etienne, E. C. Wit Statistica Neerlandica 74, 261–274
8. 2019
   Fieldwork to sample microsnails for diet and microbiome studies along the Kinabatangan River, Sabah, Malaysian Borneo K. P. Hendriks, …, F. J. Richter Mendoza, M. Schilthuizen, R. S. Etienne Malacologist 72, 33–38

Industry reports

1. 2024
   IKEA: innovative know-how for e-commerce allocation Y. Chen, M. Pellemans, F. Richter, Y. Roshandel, V. Schmeits, P. Waldert, L. Wehbi, T. C. van der Zanden Study Group Mathematics with Industry (SWI) 2024, scientific proceedings
2. 2024
   Stereo 3D cloud motion from tandem satellites Y. Zhou, S. Abrahams, F. Mendoza, M. Donà, R. Verbiest Mathematics in Industry Reports

All publications on Google Scholar

Talks

1. 2026
   How academic achievement and socioemotional learning are structurally linkedConference on Complex Systems, Binghamton
2. 2026
   The phylogenetic dark matterEuropean Meeting of Statisticians, Lugano
3. 2026
   Robust and fair prediction under distribution shiftLatin American Congress on Industrial and Applied Mathematics, Valparaíso; session organiser
4. 2025
   Neural causal regularization: extending causal invariance to deep modelsItalian Statistical Society, Genova; invited
5. 2023
   Detecting complex diversity-dependent diversification with a novel phylogenetic inference frameworkComplex Networks (CompleNet), Aveiro
6. 2023
   Phylodiversity-dependent diversification: a general frameworkWorld Conference on Natural Resource Modeling, Amsterdam
7. 2022
   Inferring the drivers of species diversification processesConference on Complex Systems, Palma de Mallorca
8. 2018
   Generalizing species diversification modelsConference on Complex Systems, Thessaloniki
9. 2018
   A statistical approach to species diversification dynamicsMathematics for Planet Earth, Utrecht; invited

Teaching

I teach from primary school to PhD level: courses at USI, thesis supervision, workshops in schools, and my own children's education at home. I write the textbooks for my courses, and lately I am turning their concepts into animations.

Now, autumn 2026

Stochastic Methods and Numerical Computing.

Earlier courses

USI

Probability and Statistics; Ordinary Differential Equations; Introduction to Data Science; Analysis of Social Networks

UTFSM

Matemáticas I

Thesis supervision

1. 2026
   Adversarial causal regularization: stress-testing distributional robustness via worst-case environment discoveryJana Muñoz RuscalledaMaster in Informatics, USI; co-supervised with Ernst Wit
2. 2025
   Causality and deep learningMetehan DündarMaster in Computational Science, USI
3. 2025
   Prediction in environmental OASI time seriesFederica MariniMaster in Statistical Sciences and Data Science, Università di Bologna
4. 2024
   Machine learning for species diversification dynamicsMattia ColbertaldoMaster in High Performance Computing Engineering, Politecnico di Milano
5. 2023
   Deep learning for parameter estimation and model classification in species diversification modelsOscar Arturo Silva CastellanosMaster in Artificial Intelligence, USI
6. 2023
   Stochastic learning for species diversification dynamicsLodovico MazzeiMaster in Computational Science, USI

Textbooks

Written for my courses: Stochastic Methods; Numerical Computing; Probability and Data Fundamentals; Irreducible Inference, on statistical inference and modelling.

Outreach

Supervision of Lavori di Maturità with the Faculty of Informatics; workshops at TecDays and at a giornata autogestita; classes at Liceo Lugano 1.

Schools and AI

My children are homeschooled, and I build their learning material with current AI tools. From that work: curriculum-aligned learning games for Chilean primary and secondary school, at juegos.whitebox.cl.

Animation

Animated explainers for the concepts in my courses, starting with Stochastic Methods.

Consulting

Contact

Write to me at richtf@usi.ch or franciscorichter@gmail.com. I work at the Faculty of Informatics, Università della Svizzera italiana, in Lugano. My code is on GitHub, and I am also on LinkedIn.