Human Machine Intelligence Research Group

Welcome to our research group page. We are supported by a variety of UKRI, H2020, and Defence & Securities funding over the years, totalling over £4.6m in PI and £16m in CI. The team is diverse and consists of 20 members from 8 countries:
 

  • Lecturer (1 Dedicated): Yang Xing - AI in autonomous systems
     

  • Research Fellows (6 Dedicated): 
    Adolfo Perrusquia (university) - reinforcement learning
    Zhuangkun Wei (EPSRC UKRI TAS-S: RS2C) - secure communications in autonomy
    Yan Zong (InnovateUK Future Flight: Fly2Plan) - secure AI in aerospace
    *recruiting summer 2021* (EPSRC) - IoT security
    *recruiting fall 2021* (EPSRC MACRO) - adversarial AI 
    Z. Du *awaiting visa* (H2020 IF) - green machine learning
     

  • Research Fellows (3 co-managed in UKRI Projects)
     

  • Ph.D. Students (7 Dedicated): 
    M. Mazzamurro (2018-22, EPSRC DTP) - scaling laws in urban science 
    Chen Li (2019-22) - green and trustworthy AI
    Schyler Sun (2019-22) - explainable and causal AI
    Mengbang Zou (2020-23) - dynamic network resilience 
    Bailu Jin (2020-23, US AFOSR) - social network influence analysis
    *recruiting fall 2021* (EPSRC TAS-S: RS2/3) - human adaptation informed secure network design
     

  • Ph.D. Students (8 Co-/External-Supervisor): 
    Andra Sonea (2018-25, EPSRC CDT) - urban digital financial service access
    Peter Strong (2020-23, EPSRC CDT) - migration modelling for refugees
    Jon Ricketts (2020-27 PT, QinetiQ) - NLP for aviation safety
    A. Abima (2020-23) - UAV communications in rural areas
    A. Dabashi (2020-23, EPSRC DTP) - smart energy infrastructure for future power systems
    J. Liu (2020-23) - AI prediction of digestate of anaerobic digestion
    P. Geragersian (2021-24, EPSRC DTP, Spirent) - AI based PNT solution for autonomous systems
    D. Holt (2021-24, EPSRC DTP, Saab) - XAI based radar signal classification for UTM

 

Research Directions

 

My directions consist of 3 main areas covering Human Machine Intelligence:
 

  1. Cyber - Autonomy and Networking in Extreme Environments (EPSRCH2020, Royal Society, DSTL, USAF, LRF)

  2. Physical - Sensing and Digital Twin of Critical Systems (EPSRC, Turing-LRFH2020, InnovateUK)

  3. Social - Human Response to a Changing Environment  (EPSRC, Turing-D&S, DSTLBritish Council, USAF)

Details of these areas and their researchers are given below.

Screenshot 2021-05-27 at 09.01.50.png

Projects: EPSRC TAS-S, EPSRC MACRO, EPSRC-PETRAS GraphSec, H2020 Mo-IoNT, H2020 GreenML for 5G, Royal Society-NSFC SmallTalk, DSTL MEDE, USAF MolSig

Cyber - Autonomy and Networking in Extreme Environments

Desig a new generation of machine learning (federated, trustworthy, tiny) and networks (radio, molecular, DNA) for extreme environments: ultra-light, energy efficient, and nanoscale. Paving the way for new 6G Internet of NanoThings and swarm UAVs connectivity, with strong application in military, healthcare, and industry.

Awards:

  • Bell Labs 2014 - Finalist

  • IEEE Best Paper 2014

  • IET Innovation Award 2015 - Winner

Key Papers:

  • "Random Sketch Learning for Deep Neural Networks in Edge Computing," Nature Computational Science, vol.1(3), Mar 2021

  • "Tabletop Molecular Communication: Text Messages Through Chemical Signals" PLOS ONE, 2013 

  • "Programmable Wireless Channel for Multi-user MIMO Transmission using Meta-surface," IEEE Globecom, 2019

  • "On the Accuracy and Efficiency of Sensing and Localization for Robotics,"
    IEEE Trans. on Mobile Computing, 2020

 

Nature Computational Science

IEEE Transactions on NanoBioScience: cover

screenshot_2017-10-05_08.52.35.png
Screenshot 2021-04-09 at 09.56.28.png
Network.png

Projects: EPSRC CoTRE, Turing DCE CHANCE, H2020 DAWN4IoE, InnovateUK ODIN, InnovateUK COCKPIT 5G, InnovateUK Fly2Plan. 

Physical - Sensing and Digital Twin of Critical Systems

Analysing and sensing the resilience of networked infrastructure as a function of local dynamics and global network topology. Minimise data collection for future Digital Twins using graph signal processing and machine learning. Inform predictive maintenance and long-term design of engineering and engineered systems.

Awards:

  • Bell Labs 2019 - Semi-Finalist

Key Papers:

  • "Sampling and Inference of Networked Dynamics using Log-Koopman Nonlinear Graph Fourier Transform,"
    IEEE Trans. on Signal Processing, 2020

  • "Network community structure of substorms using SuperMAG magnetometers," Nature Communications, vol.12, 2021

  • "Node-Level Resilience Loss in Dynamic Complex Networks," Nature Scientific Reports, 2020

  • "Optimal Sampling of Water Distribution Network Dynamics using Graph Fourier Transform,," IEEE Transactions on Network Science and Engineering, 2019

Royal Society Open Science - Cover Issue​

Nature Communications

screenshot_2017-11-02_17.40.49.png
Screenshot 2021-05-18 at 07.27.45.png
Screenshot 2019-06-19 at 22.32.41.png

Projects: Turing D&S GUARD, DSTL UNSURE, EPSRC DTG: Urban Scaling Laws, EPSRC CDT: Urban Science & Progress, British Council KEP, USAF Social Influence 

Social - Human Response to a Changing Environment 

Develop generative models and AI solutions to predict emerging events and conflict, both in the physical and cyber world. Use models to inform political science development (causal discovery, peace negotiation), and stakeholders (government, NGOs, peacekeeping, and sustainable development).

Awards:

  • Bell Labs 2016 - Semi-Finalist

Key Papers:

  • "Retool AI to Forecast and Limit Wars," Nature, 2018

  • "Common Statistical Patterns in Urban Terrorism," Royal Society OS, 2019 

  • "Simulating Imperial Dynamics and Conflict in the Ancient World," Cliodynamics, 2019

  • "Gang Confrontation: The case of Medellin (Colombia)," PLOS ONE, vol.14(12), Dec 2019

  • "Google Trends can Improve Surveillance of Type 2 Diabetes," Nature Scientific Reports, vol. 7(1), Jun 2017

Financial Times

Nature: commentary

Screenshot 2021-05-29 at 10.34.50.png
Screenshot 2021-05-18 at 07.29.53.png