Skip to content

Dr. Ken Kitson

Research Profile

I am an independent researcher and research visitor at Queen Mary University of London, specializing in causal discovery and Bayesian network structure learning. My work bridges the gap between traditional statistical methods and modern artificial intelligence to advance our understanding of causation in complex systems.

Academic Background

Current Position - Research Visitor, Queen Mary University of London - Independent Researcher, CausalIQ

Research Interests

  • Causal discovery and inference
  • Bayesian network structure learning
  • Score-based and constraint-based algorithms
  • AI-enhanced statistical methods
  • Reproducible research methodologies
  • Open science and collaborative research

Research Philosophy

I believe that understanding causation is fundamental to scientific progress and evidence-based decision making. My work focuses on developing methods that are:

  • Statistically rigorous: Grounded in solid mathematical foundations
  • Computationally efficient: Scalable to real-world datasets
  • Interpretable: Providing clear insights into causal mechanisms
  • Reproducible: Supporting open science and collaborative research
  • Practically useful: Addressing real problems across disciplines

Methodological Focus

Statistical Foundations

My research builds on established statistical theory for causal inference, including:

  • Graphical models and d-separation
  • Conditional independence testing
  • Score-based structure learning (BIC, AIC, BDeu)
  • Constraint-based algorithms (PC, FCI)
  • Hybrid approaches combining multiple paradigms

AI Integration

I explore how modern AI techniques can enhance traditional causal discovery:

  • Large language models for domain knowledge integration
  • Natural language processing for causal reasoning
  • Automated interpretation and explanation generation
  • Human-AI collaborative workflows

Reproducibility & Open Science

All research is conducted with reproducibility in mind:

  • Open source implementations of all algorithms
  • Comprehensive experimental protocols
  • Version-controlled datasets and results
  • Detailed documentation and tutorials

Collaboration & Mentoring

I'm committed to fostering the next generation of researchers in causal inference. I welcome:

  • Student collaborations: Working with graduate students on research projects
  • Academic partnerships: Joint projects with research groups
  • Industry collaboration: Applying causal methods to real-world problems
  • Open source contributions: Community-driven development of tools

Contact & Collaboration

I'm always interested in discussing research ideas and potential collaborations. Please reach out through:

  • GitHub Discussions on relevant project repositories
  • Academic conferences and workshops
  • Research seminars and invited talks

"The pursuit of causal understanding requires both mathematical rigor and creative thinking. By combining traditional statistical methods with modern AI capabilities, we can build tools that help researchers across disciplines make better causal inferences."