Defeng Liu
Polytechnique Montreal
Local AA-5505

Defeng Liu is doing his Ph.D. in Mathematics under the supervision of Prof. Andrea Lodi. His research focus on Mixed-Integer Programming, Mathematical Optimization, and Machine Learning. Namely, Defeng is interested in applying deep learning and reinforcement learning techniques in mixed integer programs to solve complex problems, aiming at bringing machine learning and operations research closer together.

Before joining the group, Defeng completed his master at CentraleSupélec (France) and holds two Computer Science and System Engineering.



  • Liu, D., Lodi, A., & Tanneau, M. (2021). Learning chordal extensions. Journal of Global Optimization, 81(1), 3-22.
  • Liu, D., Fischetti, M., & Lodi, A. (2022, June). Learning to search in local branching. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 36, No. 4, pp. 3796-3803).
  • Silva, W. A., Bobbio, F., Caye, F., Liu, D., Pepin, J., Perreault-Lafleur, C., & St-Arnaud, W. (2022). Design and Implementation of an Heuristic-Enhanced Branch-and-Bound Solver for MILP. arXiv preprint arXiv:2206.01857.
  • Liu, D., Fischetti, M., & Lodi, A. (2022). “Revisiting local branching with a machine learning lens”, Optimization Online.
  • Liu, D., Perreault, V., Hertz, A., & Lodi, A. (2022). A machine learning framework for neighbor generation in metaheuristic search. arXiv preprint arXiv:2212.11451.