cv | résumé

Here is a summary of what I have recently been doing.

Experience

  • The Eric and Wendy Schmidt Center for Data Science and Environment • DATA SCIENTIST
    2023 - present • Berkeley, CA
    Co-creating and developing open tools and solutions for environmental problem-solving.

  • UN Global Pulse • DATA SCIENCE RESEARCH FELLOW
    2022 - 2023 • New York, NY
    Multidisciplinary data science efforts for humanitarian response and situational awareness in crisis contexts. Collaborating with several UN partner entities to assess data needs and develop operational AI frameworks.

  • Vector Institute • AI SCIENTIST
    2021 - 2022 • Toronto, Canada
    Postdoctoral research work in the group of Prof. Chris J. Maddison. Designed novel deep learning models to answer research questions at the intersection of AI and optimization. Served as reference figure in optimization for students and faculty involved in the projects.

  • IBM Analytics • RESEARCH ASSISTANT
    2017 - 2018 • Montréal, Canada
    One year collaboration with the IBM‐CPLEX Optimizer development team to implement an end‐to‐end ML predictive pipeline to decide an algorithmic switch for MIQPs. Assisted scaling, integration and deployment of tool in the solver, achieving 28% runtime improvement over previous default strategy (deployed in CPLEX 12.10.0).

  • Polytechnique Montréal • TEACHING ASSISTANT
    2017 - 2019 • Montréal, Canada
    Graduate-level introduction to Mixed-Integer Linear Programming: modeling, exponential formulations, computational aspects. Exercise sessions, grading and students support (3 semesters, up to 30 students/class).

  • Polytechnique Montréal • RESEARCHER
    2015 - 2020 • Montréal, Canada
    Leading of several data science projects applying different ML techniques to new data sources. Contributed reusable methodologies to a novel research field, from data identification to careful target and feature engineering. Designed new performance metrics to account for context‐specific impact. Developed and executed automated data and analytical workflows on distributed computing clusters.

  • Business Integration Partners • CONSULTANT INTERN
    2015 • Milan, Italy
    6-months on directional-oriented projects for the client’s IT department, identifying opportunities to drive operations and innovate processes.

Education

  • Polytechnique Montréal • PH.D. in APPLIED MATHEMATICS
    2015 - 2020 • Montréal, Canada • GPA: 4.0/4.0
    Thesis: Machine learning algorithms in Mixed-Integer Programming
    Affiliations: Canada Excellence Research Chair “Data Science for real-time Decision-Making” (DS4DM), GERAD, CIRRELT
    Training: Implementation of Algorithms for Operations Research, Machine Learning, Applications of Game Theory

  • University of Padova • M.SC. in MATHEMATICS
    2012 - 2014 • Padova, Italy • 110/110 cum laude
    Training: Integer Programming, Operations Research

  • Aarhus University • ERASMUS PROGRAM
    2013 - 2014 • Aarhus, Denmark • GPA: 12.0/12.0
    Training: Combinatorics, Multi-objective Optimization

  • University of Padova • B.SC. in MATHEMATICS
    2009 - 2012 • Padova, Italy • 102/110
    Training: Linear Programming, Graph Theory

Skills

  • Data visualization + manipulation: Python, numpy, pandas, matplotlib, MS Excel, SQL
  • Machine learning + computing: scikit-learn, PyTorch, Bash, distributed computing clusters
  • Optimization + modeling: SCIP Optimization Suite, IBM ILOG CPLEX Optimization Studio, AIMMS, MATLAB
  • Languages: Italian (native), English (fluent), French (beginner)

Publications

Authors are listed alphabetically, as is standard practice in Operations Research journals and conferences, with the only exception of [1] and [3], in which authors are listed by relative contribution following the practice in Computer Science. I am first author in all reported works except [1].
For a complete list of publications, please see my Google Scholar page.

  1. Learning to Cut by Looking Ahead: Cutting Plane Selection via Imitation Learning
    Max B. Paulus, GZ, Andreas Krause, Laurent Charlin, Chris J. Maddison (2022)
    Proceedings of the 39th International Conference on Machine Learning, PMLR 162:17584‐17600. ICML 2022.

  2. A Classifier to Decide on the Linearization of Mixed-Integer Quadratic Problems in CPLEX
    Pierre Bonami, Andrea Lodi and GZ (2022)
    Operations Research.

  3. Parameterizing Branch-and-Bound Search Trees to Learn Branching Policies
    GZ, Jason Jo, Andrea Lodi and Yoshua Bengio (2021)
    Proceedings of the AAAI Conference on Artificial Intelligence, 35(5), 3931-3939. AAAI 2021.

  4. Learning MILP Resolution Outcomes Before Reaching Time-Limit
    Martina Fischetti, Andrea Lodi and GZ (2019)
    Integration of Constraint Programming, Artificial Intelligence, and Operations Research. CPAIOR 2019.

  5. Learning a Classification of Mixed-Integer Quadratic Programming Problems
    Pierre Bonami, Andrea Lodi and GZ (2018)
    Integration of Constraint Programming, Artificial Intelligence, and Operations Research. CPAIOR 2018.

  6. On learning and branching: a survey
    Andrea Lodi and GZ (2017)
    TOP 25, 207–236.

Scholarships and awards

2019 • Excellence CIRRELT Award, Doctoral Redaction Grants
2017 • GERAD Doctoral Fellowship, “Conference Fees” Competition
2017 • Honorable Mention Prize, Poster Competition, The 2017 Mixed Integer Programming Workshop

Workshops and teaching

  • Recent Advances in Integrating Machine Learning and Combinatorial Optimization
    TUTORIAL PRESENTER • AAAI 2021
    Overview of recent advances in the application of machine learning to combinatorial optimization, with special focus on Mixed-Integer Programming and the Branch-and-Bound framework.
    Watch the video (I present in Part 4)!

  • Machine Learning for Combinatorial Optimization
    TUTORIAL ORGANIZER • IJCAI 2020
    Tutorial for newcomers in the field at the intersection of ML and combinatorial optimization, to showcase the landscape of this research space using examples from the literature.

  • MTH6404 - Integer Programming
    TEACHING ASSISTANT • Polytechnique Montréal • Fall 2017, 2018, 2019
    Graduate-level introduction to various concepts in Mixed-Integer Linear Programming (MILP): modeling of combinatorial optimization problems, methods for exponential formulations, computational aspects of MILP software. Classroom exercise sessions, grading and office hours to assist students.

  • Intern Project Tutoring
    CO-TUTOR • Polytechnique Montréal • April - July 2017
    Tutoring of intern graduate student at CERC DS4DM, on the project “Exploiting variability at the root node of a branch-and-cut system”.

Selected talks and posters

  • Workshop on Data Science for Real‐Time Decision‐Making, Learning to cut by looking ahead: cutting plane selection via imitation learning. August 16, 2022.
  • Toronto AI Safety Reading Group, Verification of neural networks: a primer. June 4, 2021.
  • AAAI 2021, Parameterizing Branch-and-Bound search trees to learn branching policies. February 2-9, 2021.
  • Montréal Machine Learning and Optimization (MTLMLOpt) Seminar, Machine learning algorithms in Mixed-Integer Programming. June 26, 2020.
  • SCIP Online Workshop 2020, Parameterizing Branch-and-Bound search trees to learn branching policies. June 4, 2020. [video]
  • INFORMS Annual Meeting 2019, Reward-driven branching policies. October 20-23, 2019, Seattle, WA, USA.
  • The 2019 Mixed Integer Programming Workshop, Poster Reward-driven branching policies (G.Zarpellon, J.Jo, A.Lodi and Y.Bengio) July 15-18, 2019, Boston, MA, USA.
  • CPAIOR 2019, Learning MILP resolution outcomes before reaching time-limit. June 4-7, 2019, Thessaloniki, Greece.
  • INFORMS Annual Meeting 2018, A temporal architecture for Branch and Bound. November 4-7, 2018, Phoenix, AZ, USA.
  • ISMP 2018, Learning MILP resolution out comes before reaching time-limit. July1-6, 2018, Bordeaux, France.
  • CPAIOR 2018, Learning a classification of Mixed-Integer Quadratic Programming Problems. June 26-29, 2018, Delft, The Netherlands.
  • IBM T.J.Watson Center, Deciding whether to linearize MIQPs: a learning approach. February 8-9, 2018, Yorktown Heights, NY, USA.
  • The Aussois Combinatorial Optimization Workshop, Learning a classification of Mixed-Integer Quadratic Programming problems. January 7-12, 2018, Aussois, France.
  • Seminar “Un chercheur du GERAD vous parle!”, Learning a classification of Mixed-Integer Quadratic Programming problems. November 14, 2017, Montréal, Canada.
  • 15th EUROPT Workshop on Advances in Continuous Optimization, Learning a classification of Mixed-Integer Quadratic Programming problems. July 12-14, 2017, Montréal, Canada.
  • The 2017 Mixed Integer Programming Workshop, Poster Learning a classification of Mixed-Integer Quadratic Programming problems (P. Bonami, A. Lodi, G. Zarpellon) June 19-22, 2017, Montréal, Canada. Awarded Honorable Mention.
  • CERC DS4DM Internal Seminar, Towards learned branching decisions. March 24, 2016, Montréal, Canada.

Service

Reviewer • Mathematical Programming Computation, Computers & Operations Research, INFORMS Journal on Computing, Operations Research, CPAIOR - Integration of Constraint Programming, Artificial Intelligence, and Operations Research, AAAI - Association for the Advancement of Artificial Intelligence

Community • Girls Who Code Club organizer at Polytechnique Montréal (Winter 2020)