Maurizio Dacrema
Assistant Professor at Politecnico di Milano
Milan, Lombardy, Italy
Summary
🤩
Rockstar🎓
Top SchoolMaurizio Dacrema is an Assistant Professor at Politecnico di Milano with nine years of experience researching recommender systems evaluation and applied quantum machine learning. He progressed from PhD-with-honors to postdoc and faculty roles while teaching courses from recommender systems to applied quantum ML, and has a track record of improving reproducibility and evaluation methodology for neural recommendation models. His open-source work includes contributions to the influential RecSys 2019 evaluation repository, where he extended evaluation metrics and tooling to better assess deep learning recommenders. Beyond research, he has substantial experience in higher-education quality assurance, having served as an accreditation expert on dozens of university visits in Italy. Colleagues describe him as a methodical investigator who blends rigorous empirical critique with hands-on tooling to make research results more reliable and reusable.
9 years of coding experience
3 years of employment as a software developer
Athens certificate of participation, Electric Vehicles: the bigger picture, Athens certificate of participation, Electric Vehicles: the bigger picture at KU LeuvenDoctor of Philosophy, Computer science and engineering, PhD with honors, Doctor of Philosophy, Computer science and engineering, PhD with honors at Politecnico di Milano
Italian, English, French
Github Skills (5)
machine-learning10
python10
rep8
data-analysis8
repr8
Programming languages (2)
Github contributions (5)
MaurizioFD/RecSys2019_DeepLearning_Evaluation
Apr 2019 - Nov 2021
This is the repository of our article published in RecSys 2019 "Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches" and of several follow-up studies.
Role in this project:
Data Scientist
Data ScientistContributions:52 commits, 2 PRs, 38 pushes in 2 years 7 months
Contributions summary:Maurizio's commits primarily involve modifications to `Evaluator.py`, suggesting a focus on model evaluation within a recommender system context. Changes include the addition of new metrics like `Items_In_GT` and `Users_In_GT`, and the refinement of existing evaluation methods. These actions point towards an effort to improve the evaluation framework for recommender systems, likely to assess the performance of deep learning models or other recommendation algorithms detailed in the article.
bprmffollowhybrid-recommender-systemcontent-based-recommendationrecommendation
MaurizioFD/spotify-recsys-challenge
Oct 2018 - Nov 2020
A complete set of Recommender Systems techniques used in the Spotify Recsys Challenge 2018 developed by a team of MSc students in Politecnico di Milano.
Contributions:4 commits, 2 PRs, 4 pushes in 2 years 1 month
milanopolitecnico-di-milanopythonrecommendertechniques
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.