Rail 05

Machine Learning Foundations

Fundamental research in deep learning architectures, reinforcement learning, and the mathematical foundations of intelligence — pushing the theoretical boundaries of AI.

Machine Learning Foundations

Machine Learning Foundations

Our machine learning foundations group advances the theoretical and practical understanding of deep learning, reinforcement learning, and neural architecture design. From meta-learning and continual learning to federated systems and causal inference, we tackle the open problems that underpin real-world AI deployment.

Deep Reinforcement Learning Meta-Learning NAS Causal Inference Federated Learning Continual Learning

Key Research Topics

Deep Learning Architecture

Neural architecture search, attention mechanisms, and scalable training strategies for large-scale deep learning models.

Reinforcement Learning

Deep RL algorithms for sequential decision-making in partially observable, multi-agent, and non-stationary environments.

Federated & Distributed AI

Privacy-preserving distributed learning frameworks that train models across decentralised nodes without raw data sharing.

Explainability & Fairness

Interpretable model design, algorithmic fairness auditing, and causal attribution methods for high-stakes AI applications.

Current Projects

Ml · Active · 2024–2026

MetaAdapt: Cross-Domain Few-Shot Learning

Cross-domain few-shot learning framework that rapidly adapts AI models to new tasks with minimal labelled data.

MAML ProtoNets Hugging Face PyTorch Ray Tune
Ml · Active · 2025–2027

FairML: Algorithmic Fairness Auditing Suite

Algorithmic fairness auditing suite that detects and mitigates bias in machine learning systems across sensitive domains.

Fairlearn SHAP CausalML Python Streamlit

Related Publications

Interested in Collaborating?

We welcome partnerships with institutions, NGOs, and researchers working at the frontier of AI and this domain.