People

INFORMED-AI is an inclusive, evolving research programme, building collaborations between information theorists and researchers in AI, control and robotics across the UK. We are also expanding our network with affiliates from other institutions and past members who have gone on to their next career move.  

Trustworthy Cooperation in Heterogeneous Teams

Jonathan Lawry

Research interests

  • Social learning
  • Collective decision making
  • Approximate reasoning

Hub role

Work package: Cooperation in Heterogeneous Teams

Jonathan Lawry

Jianhong Wang

Research interests

  • Multi-Agent Reinforcement Learning
  • Cooperative Game Theory (coalition formation and credit assignment)
  • Multi-Agent System modelling for real-world problems (e.g., smart grids, robotics, etc.)

Hub role

Jianhong Wang

Yiannis Demiris

Research interests

  • Assistive Robotics and Human–Robot Interaction
  • Human action understanding and human-centred robot behaviour adaptation
  • Multimodal learning and fusion
  • Distributed learning and in particular human-in-the-loop learning with privacy

Hub role

Work package: Cooperation in Heterogeneous Teams

Yiannis Demiris

Fernando Estévez Casado

Research interests

  • Distributed/federated learning and privacy in robotics
  • Assistive robots and trustworthy HRI (focus on mobility aids)
  • User-centred robot learning, personalisation, and adaptation over time
  • Deployment on real-world applications, platforms, and human participants

Hub role

Fernando Estévez Casado

Amanda Prorok

Research interests

  • Collective intelligence
  • Robotics
  • Multi-agent and muti-robot systems
  • Applications such as:
    • Automated transport and logistics
    • Environmental monitoring
    • Surveillance
    • Search

Hub role

Work package: Cooperation in Heterogeneous Teams

Amanda Prorok

Maxence Boels

Research interests

  • Robotics
  • Multi-agent ariel robotics

 

Hub role

Work package: Cooperation in Heterogeneous Teams

Photo of Maxence Boels

Learning to Compress and Communicate

Ioannis Kontoyiannis

Research interests

Information theory

  • Communication and compression at pragmatic rates
  • Truly high-SNR channel capacity
  • Compression of truly sparse data

Statistics

  • Bayesian causal discovery
  • Regression and estimation for the Levy state space model

Machine learning and AI

  • Bayesian/MDL neural network inference
  • Sparse Gaussian processes
  • Design and analysis of efficient Reinforcement Learning
  • Stochastic Approximation algorithms

Hub role

Work package: Learning to Compress and Communicate

Ioannis Kontoyiannis

Lampros Gavalakis

Research interests

  • Information theory
  • Fundamental limits
  • Entropy
  • Inequalities
  • De Finetti’s theorem
  • Sampling bounds
  • Entropic CLT and applications
  • Entropy of Gaussian mixtures and applications

Hub role

Lampros Gavalakis

Deniz Gündüz

Research interests: Information Processing and Communications Laboratory (IPC-Lab)

Machine learning (ML) for data compression and communication

  • ML-based channel code design
  • ML-based image/video compression
  • ML-based channel estimation and channel state compression

Distributed/ federated learning (FL)

  • Communication-efficient FL
  • Over-the-air computation for FL

Semantic communications

  • Deep–learning aided wireless video transmission and intelligence

Privacy and security in learning

  • Byzantine attacks in federated learning
  • Private data sharing

Hub role

Work package: Learning to Compress and Communicate

Deniz Gündüz

Yassine Hamdi

Research interests:

  • lossy data compression and information theory
  • rate-distortion-perception theory
  • practical lossy compression codecs
  • channel simulation
  • distributed coordination
  • theory of diffusion models

Hub role

Photo of Yassine Hamdi

Pier Luigi Dragotti

Research interests:

  • Mathematical methods for signal representation and for high-dimensional data analysis​

  • Model-based deep learning ​

  • Applications in inverse problem, image/video compression and transmission​

Hub role

Work package: Learning to Compress and Communicate

Pier Luigi Dragotti

Di You

Research interests:

  • signal processing and machine learning,

in particular:   

  • inverse problems
  • deep joint source-channel coding,
  • diffusion models.

Hub role

Work package: Learning to Compress and Communicate

Photo of Di You

Oliver Johnson 

Research interests

  • Information theory fundamentals​

  • Sparse inference​

Hub roles

  • Work package: Learning to Compress and Communicate
Oliver Johnson

AI Theory for Quantum Information and Computing

Nilanjana Datta

Research interests

  • Quantum entropies and divergences
  • Quantum hypothesis testing
  • Entanglement theory
  • Quantum communication

Hub role

Work package: Quantum Information and Computing

Nilanjana Datta

Bjarne Bergh

Research interests

  • Quantum hypothesis testing
  • Quantum machine learning
  • Distributed quantum algorithms

Hub role

Bjarne Bergh

Neil Walton

Research interests

  • Queueing
  • Congestion
  • Networks
  • Probability
  • Optimization

Hub role

Work package: Quantum Information and Computing

Neil Walton

Thiru Vasantam

Research interests

  • Applied probability and queueing theory
  • Internet of Things
  • Performance modeling and analysis of communication networks
  • Quantum networking
  • Sequential analysis and changepoint detection

Hub role

Work package: Quantum Information and Computing

Thiru Vasantam

Decentralised Learning Under Information Constraints

Ayalvadi Ganesh

Research interests

Random graphs and processes

  • Connectivity
  • Epidemics
  • Opinion dynamics

Decentralised algorithms

  • Information dissemination
  • Collective decision-making and coordination

Stochastic optimisation

  • Resource allocation
  • Load balancing
  • Multi-armed bandits

Hub roles

Ayalvadi Ganesh

Guru Ganeshan

Research interests

  • Using probabilistic and algebraic techniques to design and analyze codes/methodologies for smart data storage/retrieval with feature selection
  • Analysis of classification and dynamic evolution; for example: random forest classifier,  under independent and identically distributed (i.i.d.) conditions

Hub role

Guru Ganeshan

Po-Ling Loh

Research interests

  • Theoretical statistics
  • Random graphs and networks
  • Robustness
  • Differential privacy

Hub role

Po-Ling Loh

Paul Rosa

Research interests

  • Theoretical foundations of nonparametric and semiparametric statistical techniques
  • Bayesian approaches
  • Functional analysis
  • Spectral theory
  • Differential geometry.

Hub role

Paul Rosa

Sidharth Jaggi

Research interests

  • Adversarial information-processing
  • Stealthy/covert communication
  • Group testing
  • Network Coding

Hub roles

Sidharth Jaggi