Hi, I’m Haozhe Tian, a PhD student at the Dyson School of Design Engineering, Imperial College London, UK. My research focuses on Reinforcement Learning (RL), with applications in:

  • Safety-critical systems, such as the management of non-communicable diseases including diabetes and cardiovascular conditions,
  • NP-hard combinatorial optimization, such as identifying structurally significant nodes in large-scale complex networks with millions of nodes.
Before starting my doctorate, I completed an MSc (Distinction) in Communications and Signal Processing at Imperial and a BEng in Pattern Recognition and Control at Beihang University, China. I also spent time as a research assistant at the ASTAPLE Lab at Hong Kong Polytechnic University.

Haozhe Tian

haozhe dot tian21 at imperial dot ac dot uk

PhD student at Dyson School of Design Engineering, Imperial College London, UK

News
  • [2024 Oct] One paper "Reinforcement Learning with Adaptive Regularization for Safe Control of Critical Systems" accepted by NeurIPS 2024.
  • [2024 Aug] I received Imperial I-X Moonshot Seed Funding to work on "Safe Reinforcement Learning for Critical Applications".
  • [2023 Oct] I began my PhD as part of the EU CoDiet project.
  • [2022 Oct] I was awarded the Ivor Tupper Prize for Excellence in Signal Processing, Broadcast and Video Technology for my Master thesis.

Publications

Press blocks for full-text Conference Journal Preprints
Also available on Google Scholar.

  • arXiv Machine Intelligence on the Edge: Interpretable Cardiac Pattern Localisation Using Reinforcement Learning
    Haozhe Tian, Qiyu Rao, Nina Moutonnet, Pietro Ferraro, and Danilo Mandic
  • arXiv Learning Network Dismantling without Handcrafted Inputs
    Haozhe Tian, Pietro Ferraro, Robert Shorten, Mahdi Jalili, and Homayoun Hamedmoghadam
  • NeurIPS 2024 Reinforcement Learning with Adaptive Regularization for Safe Control of Critical Systems
    Haozhe Tian, Homayoun Hamedmoghadam, Robert Shorten, and Pietro Ferraro
  • IEEE BigData 2023 CGP: Centroid-guided Graph Poisoning for Link Inference Attacks in Graph Neural Networks
    Haozhe Tian, Haibo Hu, and Qingqing Ye
  • EMBC 2023 Hearables: Heart Rate Variability from Ear Electrocardiogram and Ear Photoplethysmogram (Ear-ECG and Ear-PPG)
    Haozhe Tian, Edoardo Occhipinti, Amir Nassibi, and Danilo Mandic
  • arXiv Mission-Aligned Learning-Informed Control of Autonomous Systems: Formulation and Foundations
    Vyacheslav Kungurtsev, Gustav Sir, Akhil Anand, Sebastien Gros, Haozhe Tian, and Homayoun Hamedmoghadam
  • IEEE TIM Instrumentation of surface plasmon microscopy: complete scheme of signal extractions
    Bei Zhang, Haozhe Tian, Tianyu Xiao, and Jing Zhang

Photos