Konstantinos Barmpas

Konstantinos Barmpas is a Postdoctoral Research Associate at Imperial College London's Department of Computing within the Faculty of Engineering. His research focuses on developing Generative Foundation Models for biosignals in Brain-Computer Interfaces (BCIs), emphasizing causal reasoning and deep learning applications. He holds a PhD in Computer Science (AI) from Imperial College London and an MEng from Imperial College London/ETH Zurich.

• Education:
  • PhD in Computer Science (Imperial College London, 2020–2024)
  • MEng in Electrical & Electronic Engineering (Imperial College London, 2016–2020)
  • Master Year Abroad at ETH Zurich (2019–2020)

His research bridges deep learning with BCIs, addressing topics like differentiable signal processing, causal modeling, and ethical challenges in GenAI-powered neural interfaces. He leads projects like MyBrainCommands and MyBrainStates, approved by Imperial's Research Governance team. His work has been featured in outlets like Sky News, BBC, and The Times.

Barmpas contributes extensively to academia: he co-organized the London Geometry and Machine Learning (LOGML) Summer School, led the Imperial Computing Conference (ICC), and served on Imperial's Equity, Diversity, and Culture Committee. He reviews for top AI journals/conferences (e.g., IEEE Transactions, NeurIPS workshops) and mentors students through initiatives like Microsoft Student Ambassador programs. He is a Fellow of the Higher Education Academy (FHEA) and an ELLIS Society member.

His innovations include the CogiGame EEG-VR platform for stroke rehabilitation and causal frameworks for brainwave modeling. Over 15 peer-reviewed publications span topics like EEG signal processing, neural network geometry, and BCI ethics. Patents include classification of brain activity signals (GB2605270).