Hello, I’m Niki Triantafyllou

I’m a PhD candidate at Imperial College London

I work with Dr. Maria Papathanasiou in the Sargent Center for Process Systems Engineering.

My research lies at the intersection of machine learning, optimization, and control, with a focus on building optimization-aware learning systems that are scalable and interpretable. I develop hybrid algorithms that integrate learning and decision-making, including:

  • Enhancing combinatorial optimization with neural architectures and energy-based learning
  • Embedding differentiable optimization layers into neural networks
  • Designing model-based and sample-based Bayesian optimization frameworks—often I combine them in multi-fidelity settings—for efficient high-dimensional, constrained black-box problems

Motivated by real-world challenges in next-generation personalized cell and gene therapies and energy systems, I use domain-specific insight to build intelligent decision-making systems that can plan, adapt, and reason under uncertainty.

My research is supported by the Marit Mohn PhD Scholarship.

🏆 My MSc thesis on bilevel decomposition of large-scale mixed integer optimization models for personalized medicine supply chains earned the Best Poster Award at ESCAPE32 and the Best Research Project Prize in the MSc in Advanced Chemical Engineering at Imperial College London.

I was an Applied Scientist Intern at Amazon from September 2024 to March 2025.

Research Interests

  • • Learning for Optimization
  • • Hybrid Modeling
  • • Bayesian Optimization
  • • Model Predictive Control
  • • Mixed Integer Optimization
  • • Graph Neural Networks
  • • Gaussian Processes
  • • Reinforcement Learning