Nathalie Kirch

Nathalie Kirch

I am a PhD student in Computer Science Research at Imperial and King’s College London. My research mainly focuses on mechanistic interpretability and robustness in LLMs. I have a broad interdisciplinary background in cognitive psychology, philosophy, and...

Rui Wang

My research topic is ‘Interpretable Spatio-Temporal Crime Prediction Models Based on Machine Learning’. This study aims to develop interpretable spatio-temporal crime prediction models in London. Current crime prediction models struggle to balance...

Omar Adalat

Ensuring that reinforcement learning emits policies that are safe and verifiable is crucial, and my project aims to do so through integrating Formal Methods, such as using temporal logic to shield the policy from performing unsafe action or synthetising reward...

Dylan Garcia

My research investigates how AI and digital literacy can be more effectively taught, measured, and co-designed with marginalized communities, focusing on first-generation Hispanic immigrants in New York City. It will evaluate current digital literacy programs and...

Christos Ziakas

I’m doing my PhD at Imperial on learning from action-free videos, adapted for unseen and open-ended environments by leveraging generative AI. My research focuses on enhancing learning from expert demonstrations in a scalable and interpretable way, contributing...
Israel Mason-Williams

Israel Mason-Williams

My background is in the theory of deep learning, and during my PhD, I aim to use ideas of functional similarity and mechanistic interpretability to better understand neural networks and their outputs from a foundational perspective. My previous research has focused on...

Lukas Twist

My research interests lie at the intersection between software engineering and AI, aiming to utilise my experience in the software engineering industry to contribute to the creation of robust and innovative AI systems. I’m currently researching problems that arise...
Akchunya Chanchal

Akchunya Chanchal

My PhD focuses on using concepts from Causality (more specifically, Actual Causality) to prove bounds and improve learnt representations (weights) of deep learning models, so that they learn concepts, rather than patterns, allowing for better generalisation of models...
Gerrit Quaremba

Gerrit Quaremba

I am a PhD student working on the collaborative content creation use case as part of the UKRI research project PHAWM (Participatory Harm Auditing Workbenches and Methodologies), which is aligned with the UKRI Centre for Doctoral Training in Safe and Trusted AI. My...

Katharine Sparks

The primary aim of my research is to leverage the respective strengths of knowledge graph reasoners and large language models. While LLMs can efficiently generate highly expressive natural language output, their probabilistic nature precludes consistency as a...
Ma’ayan Armony

Ma’ayan Armony

My research focuses on leveraging common-sense knowledge for the process of robot task planning. More specifically, it aims to improve the safety and trustworthiness of plans (sequences of actions) that are generated for a robot to execute, by using structured...
Basak Kocaoglu

Basak Kocaoglu

My research is focused on finding the loopholes that give rise to specification gaming or reward hacking problems in AI systems. I am working on a mathematical logic framework where we can formalise multiple redundant causal paths contributing to such outcomes, and...