Dimitar Georgiev

Dimitar Georgiev

PhD student in AI & ML at Imperial College London. Academic visitor at University of Oxford.

I am a PhD student in Artificial Intelligence and Machine Learning at Imperial College London working on deep learning approaches for analytical chemistry, chemometrics and drug discovery under the supervision of Prof. Mauricio Barahona and Prof. Dame Molly Stevens.


Interests

  • Deep learning
  • ML for chemometrics
  • ML for molecules
  • AI for science
  • AI for healthcare
  • Software

Education

  • PhD in AI & ML

    Imperial College London, UK

    Oct 2021 - Present


  • MMath in Mathematics

    University of Southampton, UK

    Sep 2017 - Jul 2021


Recent highlights

2024:
    November:
    • 🪧 Poster presentation at RamanFest2024 - won Best PhD Poster award

      • Title: Hyperspectral unmixing for Raman spectroscopy via physics-constrained autoencoders

    October:
    • 📄 Paper published in PNAS

      • Title: Hyperspectral unmixing for Raman spectroscopy via physics-constrained autoencoders

    September:
    • 🎤 Oral presentation at BioMedEng24

      • Title: Hyperspectral Unmixing for Raman Spectroscopy via Physics-Constrained Autoencoders
      📍London, UK

    • 🪧 Poster presentation at BioMedEng24

      • Title: RamanSPy: An Open-Source Python Package for Raman Spectroscopy Data Analysis
      📍London, UK

    July:
    • 🎤 Oral presentation at AI4Science workshop at ICML'24

      • Title: Hyperspectral Unmixing for Raman Spectroscopy via Physics-Constrained Autoencoders
      📍Vienna, Austria

    • 🪧 Poster presentation at AI4Science workshop at ICML'24

      • Title: RamanSPy: Augmenting Raman spectroscopy data analysis with AI
      📍Vienna, Austria

    • 🪧 Poster presentation at ML4LMS workshop at ICML'24

      • Title: Hyperspectral Unmixing for Raman Spectroscopy via Physics-Constrained Autoencoders
      📍Vienna, Austria

    • 🪧 Poster presentation at ML4LMS workshop at ICML'24

      • Title: RamanSPy: Augmenting Raman spectroscopy data analysis with AI
      📍Vienna, Austria

    June:
    • 🪧 Poster presentation at MoML'24

      • Title: Accelerating Molecular Graph Neural Networks via Knowledge Distillation
      📍Montreal, Canada

    • 🏫 Participated in the ML for Drug Discovery Summer School

      • 📍Montreal, Canada

    May:
    • 📄 Paper published in Analytical Chemistry

      • Title: RamanSPy: An Open-Source Python Package for Integrative Raman Spectroscopy Data Analysis

    • 🪧 Poster presentation at CAI4H'24

      • Title: Hyperspectral Unmixing for Raman Spectroscopy via Physics-Constrained Autoencoders
      📍Edinburgh, UK

    March:
    • 🏛️ Joined University of Oxford as Academic Visitor

      • 📍Oxford, UK

    • 📝 New preprint is out

      • Title: Hyperspectral Unmixing for Raman Spectroscopy via Physics-Constrained Autoencoders

2023:
    December:
    • 🪧 Poster presentation at NeurIPS'23

      • Title: Accelerating Molecular Graph Neural Networks via Knowledge Distillation
      📍New Orleans, USA

    November:
    • 🎤 Oral presentation at LoG'23

      • Title: Accelerating Molecular Graph Neural Networks via Knowledge Distillation
      📍virtual

    September:
    • 📄 Paper accepted to NeurIPS'23

      • Title: Accelerating Molecular Graph Neural Networks via Knowledge Distillation

    July:
    • 🪧 Poster presentation at SynS&ML workshop at ICML'23

      • Title: Accelerating Molecular Graph Neural Networks via Knowledge Distillation
      📍Hawaii, USA

    • 📝 New preprint is out

      • Title: RamanSPy: An Open-Source Python Package for Integrative Raman Spectroscopy Data Analysis

    June:
    • 💻 New Python package called RamanSPy

      • Scope: Raman spectroscopy data analytics

    • 📝 New preprint is out

      • Title: Accelerating Molecular Graph Neural Networks via Knowledge Distillation

    May:
    • 🪧 Poster presentation at CAI4H'23

      • Title: Cell phenotyping via hyperspectral unmixing autoencoders
      📍York, UK


Selected publications

Hyperspectral unmixing for Raman spectroscopy via physics-constrained autoencoders

Dimitar Georgiev, Álvaro Fernández-Galiana, Simon Vilms Pedersen, Georgios Papadopoulos, Ruoxiao Xie, Molly M. Stevens, Mauricio Barahona

Proceedings of the National Academy of Sciences of the United States of America (PNAS), 2024

Code

This work has also been presented at:
  • AI for Science Workshop at International Conference on Machine Learning (oral), 2024
  • Machine Learning for Life and Material Science Workshop at International Conference on Machine Learning, 2024
  • BioMedEng24 Conference (oral), 2024
  • RamanFest Conference (Best PhD Poster award), 2024
RamanSPy: An open-source Python package for integrative Raman spectroscopy data analysis

Dimitar Georgiev, Simon Vilms Pedersen, Ruoxiao Xie, Álvaro Fernández-Galiana, Molly M. Stevens, Mauricio Barahona

Analytical Chemistry, 2024

Code Star

This work has also been presented at:
  • AI for Science Workshop at International Conference on Machine Learning, 2024
  • Machine Learning for Life and Material Science Workshop at International Conference on Machine Learning, 2024
  • BioMedEng24 Conference, 2024
Accelerating Molecular Graph Neural Networks via Knowledge Distillation

Filip Ekström Kelvinius*, Dimitar Georgiev*, Artur Petrov Toshev*, Johannes Gasteiger

*Equal contribution. Order was determined by rolling a dice.

Conference on Neural Information Processing Systems (NeurIPS), 2023

Code

This work has also been presented at:
  • Synergy of Scientific and Machine Learning Modeling Workshop at International Conference on Machine Learning, 2023
  • Learning on Graphs Conference (oral), 2023
  • Molecular Machine Learning Conference, 2024

My complete, up-to-date publication record is available on Google Scholar.