Valerio Siniscalco
valerio@siniscalco.me codeberg linkedinI finished a BSc in Computer Science at the University of Milan this summer, and in September I start an MSc in IT and Cognition at the University of Copenhagen, where I am now based.
Most recently, I spent seven months at the Centre for Mind/Brain Sciences in Trento on topographic regularisation in convolutional networks, spanning end-to-end training, frozen-backbone adaptation, ensembling, and meta-learning, analysed for representational similarity, calibration, and adversarial robustness. I am co-authoring a paper on it with the group.
Before that, I was at ITU Copenhagen’s NERDS Lab, building an NLP and network-science pipeline for my BSc thesis, entity linking and statistical backboning across five Wikipedia language editions and millions of edges, to study how connectivity differs by gender, origin, and era. The dataset is published on Zenodo and became a case study for a paper, currently under review.
Timeline
-
2026 –
MSc, IT and Cognition current University of Copenhagen, Faculty of Humanities Applied machine learning, language technology and computer vision coursework, covering the technologies behind large language models alongside cognitive science, from the Faculty of Humanities.
-
Jan – Jul 2026
Research Intern Centre for Mind/Brain Sciences (CIMeC), University of Trento, supervised by Prof. Uri Hasson Topographic regularisation in convolutional networks, implemented in PyTorch, across end-to-end training, frozen-backbone adaptation, ensembling, and meta-learning. Analysed for representational similarity, calibration, and adversarial robustness.
-
Sep – Dec 2025
Research Intern Networks, Data, and Society (NERDS) Lab, IT University of Copenhagen, supervised by Prof. Michele Coscia How historical figures are connected across five Wikipedia language editions, and how that structure differs by gender, origin, and era. Built an NLP and network-science pipeline end to end, from entity linking to statistical backboning, over millions of edges. Supported by the Danish Data Science Academy.
-
2025 – 2026
Co-Founder, Polyneta Smart-cities startup, Milan Privacy-preserving machine learning and NLP prototypes for municipal data analysis. University Startup Challenge winner, out of 400 teams.
-
2022 – 2026
BSc, Computer Science Università degli Studi di Milano 110/110 cum laude. Thesis: Representational Bias on Wikipedia: A Multilingual, Multi-Attribute Network Analysis, written at ITU Copenhagen’s NERDS Lab.
Projects
- wiki-network-backboning Builds the multilayer Wikipedia biography network from the Cross-Verified Database of Notable People, with edge weights from entity-linking-based mention detection in body text.
- Multilingual Mention-Weighted Network of Wikipedia Biographies The dataset built by the pipeline above, a directed multilayer network of Wikipedia biographies in five languages, aligned across editions by Wikidata ID and published on Zenodo.