Mengjia Zhu

Mengjia Zhu

PDRA at UoM. My current research interests include black-box optimization and control, with particular emphasis on active, preference, and safe learning.

Welcome to my homepage!


I’m Mengjia (梦佳), a Post-Doctoral Research Associate (PDRA) at University of Manchester under the supervision of Dr. Dongda Zhang.

I was a PhD student in Systems Science in the DYSCO research unit supervised by Prof. Alberto Bemporad at the IMT School for Advanced Studies Lucca, Italy. I received my bachelor degree in Chemical Engineering from the University of Washington, Seattle, WA, USA, in 2017 and my master degree in Advanced Chemical Engineering with Process System Engineering from the Imperial College London, UK, in 2019 under the supersivion of Prof. Sandro Macchietto.

My current research interests include black-box optimization and control, with particular emphasis on active, preference, and safe learning.

Updates:

  • [Oct-22-2024] Our paper “Discrete and mixed-variable experimental design with surrogate-based approach” has been accpeted by Digital Discovery.
  • [May-30-2024] I successfully defended my thesis :)
  • [April-23-2024] Check out our paper “Discrete and mixed-variable experimental design with surrogate-based approach” on ChemRxiv

  • [March-04-2024] I am starting my PDRA position at UoM under the supervision of Dr. Dongda Zhang. The PDRA is funded by the following EPSRC Grant: Artificial Intelligence Enabling Future Optimal Flexible Biogas Production for Net-Zero (EP/Y005600/1).

  • [Sept-28-2023] Presented our paper “Learning critical scenarios in feedback control systems for automated driving” at ITSC 2023, slides

  • [August-23-2023] I am starting my visiting period at the Optimisation and Machine Learning for Process Engineering group under the supervision of Dr. Ehecatl Antonio del Río Chanona in the Department of Chemical Engineering at Imperial College London. Thanks for hosting me here!

  • [July-13-2023] Our paper “Learning critical scenarios in feedback control systems for automated driving” has been accpeted by ITSC 2023, also available on arXiv

  • [April-25-2023] Our paper “Multi-agent active learning for distributed black-box optimization” has been accepted by IEEE Control Systems Letters.
  • [March-05-2023] Our paper “Specification-guided critical scenario identification for automated driving” is available online now, also available on arXiv
    • Adam will present it at the 25th International Symposium on Formal Methods (FM 23) on March 8th, link

  • [March-03-2023] Check out the updated MATLAB version of GLIS

  • [Feb-10-2023] Check out our paper “Global and Preference-based Optimization with Mixed Variables using Piecewise Affine Surrogates “ on arXiv
    • Code is availalbe at PWAS

  • [Jan-23-2023] The GLIS package is now available via pip install! Check it out at GLIS

  • [Sept-26-2022] Check out our paper “Learning Critical Scenarios in Feedback Control Systems for Automated Driving” on arXiv

  • [Sept-2022] C-GLISp paper is published in IEEE TCST