Mateja Jamnik

Researcher in Artificial Intelligence

Group

PhD Students

  • Urska Matjasec: Interpretable methodology with high performance for tabular data.
  • Julia Komorowska: Dynamics-informed protein generative modelling. (First supevisor with Pietro Lio)
  • Xiangjian Jiang: Interpretable machine learning models for low-sample-size regimes in healthcare.
  • Oscar Hill: Concept Discovery for Explainable AI.
  • Yael Konforti: Explainable AI and bias.
  • Marco Dos Santos: LLMs for mathematical reasoning.
  • Haoyan Luo: Explainability of large language models.
  • Yuval Shalev: Training Language Agents to Learn from Experience.
  • George Shaikovski: Explainable and Steerable Complex Reasoning in Transformer-based Language Models.
  • Elaf Almahmoud: The Value of AI Assistance. (First supervisor with Umang Bhatt)
  • Maleakhi Wijaya: Interpretability of LLMs for Safety.
  • Sid Mamidanna: Understanding Latent Communication Channels in Same-model Agent Swarms.

Past PhD Students

  • Konstantin Hemker (2026): Multimodal Representation Learning for Spatial Biology: Fusion, Merging, Alignment, and Dynamic.
  • Mateo Espinosa Zarlenga (2025): Help Wanted: Robust Concept Interventions for Interpretable Deep Neural Networks.
  • Andrei Margeloiu (2025): Breaking the curse of dimensionality in low-data tasks.
  • Dimitrios Deslis (2025): Developing an Interactive Computer Support System to Facilitate Teacher Learning of Proof-related Instruction in Mathematics. (First supervisor with Andreas Stylianides in Education Faculty)
  • Albert Qiaochu Jiang (2024): Language models for verifiable mathematical automation: Interaction, integration, and autoformalization.
  • Mei Yang (2024): Designing a Computer-assisted System For Prospective Mathematical Teachers’ Training of Proof-Related Instruction . (First supervisor with Andreas Stylianides in Education Faculty)
  • Dmitry Kazhdan (2024): Enhancing Interpretability: The Role of Concept-based Explanations Across Data Types. (First supervisor with Pietro Lio)
  • Paul Scherer (2023): Distributional and relational inductive biases for graph representation learning in biomedicine. (First supervisor with Pietro Lio)
  • Agnieszka Slowik (2023): Out-of-distribution generalisation in machine learning. (First supervisor with Sean Holden)
  • Aaron Stockdill (2021): Automating representation change across domains for reasoning.
  • Edward Ayers (2021): A Tool for Producing Verified, Explainable Proofs. (First supervisor with Tim Gowers in Maths Department)
  • Botty Dimanov (2021): Interpretable Deep Learning: Beyond Feature-Importance with Concept-based Explanations.
  • Duo Wang (2021): Neural Diagrammatic Reasoning. (First supervisor with Pietro Lio)
  • Advait Sarkar (2017): Interactive Analytical Modelling. (First supervisor with Alan Blackwell and Martin Spott)
  • Matej Urbas (2013): Mechanising Heterogeneous Reasoning in Theorem Provers.
  • Daniel Winterstein (2004): Automating Diagrammatic Reasoning in Continuous Domains. (Second supervisor with Alan Bundy at University of Edinburgh)

Past Post Doctoral Researchers