Welcome to Chi Zhang's Homepage

I am a PhD student at School of Computer Science, the University of Exeter, supervised by Dr. Wenjie Ruan and Prof. Zhongdong Wang. Before that, I received my Dipl.-Ing. (equals Master) Mechatronical Engineering at the Technical University of Dresden, and my BEng Mechatronical Engineering from Zhejiang University.

Research Interest

My research focuses on robust deep learning model and its applications on various safety-critical systems.

Despite the wide use and huge success of deep learning in recent years, deep learning models are demonstrated to be vulnerable and not robust to environmental uncertainty and adversarial perturbed noisy data. The primary goal of my research will be developing robust deep learning models including convolutional and recurrent neural networks, so that the trained deep learning models are robust and remain highly accurate under any environmental uncertainty such as adversarial perturbation, data missing, and data pollution.

We envision that any systems with deep learning components will benet from this research, especially those safety-critical missions operated by semi- or fully-autonomous systems, such as self-driving cars, automated rescuing robots, automated medical analysis/diagnosis.

This research will also enable the deep learning models, beyond the computer science community, to be applicable on a wider range of applications, such as various prediction, classification, and regression problems in industrial systems where the data is usually noisy, polluted, or/and missing. At the final stage of this research, besides the theoretical and algorithmic contribution to robust deep learning, several case studies from autonomous systems and industrial systems will be explored and presented.

News & Facts

Continue my PhD study at the University of Exeter

Start my PhD study at Lancaster University

Graduated from Technical University of Dresden, awarded Dipl.-Ing.
Thesis: Topology Optimization of an Electromagnetic Valve Actuator with Level-Set and Adjoint Variable Method
(Topologieoptimierung eines elektromagnetischenVentilaktors mittels Level-Set durch die Adjoint Variable Method)

6 month Internship at TP-Link Deutschland GmbH in Frankfurt

Academia Services

External Reviewer

NeurIPS 2020
CIKM 2020
ECML 2020
IJCAI 2020
ICML 2020

Skills & Hobbies

Chinese is my mother language. I can also speak English and German fluently.

I have experience of coding with matlab and python. I am also familiar with mechanical and electrical design tools such as Flecs, Femm, Ansys, Audocad, Comsol...

My hobbies are swimming and table tennis.