Academic Profile


Raghavendra Selvan (Raghav) is currently an Assistant Professor at the University of Copenhagen, with joint responsibilities at the Machine Learning (ML) Section (Dept. of Computer Science), Kiehn Lab (Department of Neuroscience) and the Data Science Laboratory. He received his PhD in Medical Image Analysis (University of Copenhagen, 2018), his MSc degree in Communication Engineering in 2015 (Chalmers University, Sweden) and his Bachelor degree in Electronics and Communication Engineering degree in 2009 (BMS Institute of Technology, India). Raghavendra Selvan was born in Bangalore, India.

His current research interests are broadly pertaining Resource Efficient ML, Medical Image Analysis with ML, Quantum Tensor Networks and Graph Neural Networks. Of late, another overarching theme of his research interests lie at the intersection of sustainability and ML where he is interested in investigating sustainability with ML, and also the sustainability of ML.


  • 2023-05: RS has been nominated for the UCPH Innovation Award 2023 for contributing to develop Carbontracker.
  • 2023-05: RS will be one of the coordinators of the Green Transition and AI thematic solution center at UCPH
  • 2023-02: Check out the ongoing reading group on Sustainable Machine Learning hosted by us here.
  • 2023-04: Contributed to a popular news article in Süddeutsche Zeitung on the carbon footprint of large language models.
  • 2023-01: Funding for one PhD student to study fronto-temporal dementia with AI at St. Andrews University as co-applicant
  • 2023-01: Paper accepted to Digital Discovery
  • 2022-12: Dept. of Computer Science Dissemination Award and ML Section award for contributions in 2022.
  • 2022-12: Copenhagen Summer University course on Climate-friendly AI approved to be delivered in W34
  • 2022-11: Paper accepted to be presented at 6th Northern Lights Deep Learning Conference (NLDL).
  • 2022-10: UK Research & Innovation grant for Environmental sustainability in Life Sciences as co-applicant (~890k DKK)
  • 2022-09: Paper accepted to NPJ Computational Materials.

Research Projects

(Switching to first person :)

I am actively involved in a wide range of research projects as part of my roles at the Machine Learning section and Data Science Lab of University of Copenhagen. Brief descriptions of ongoing and completed projects are listed below.

Ongoing projects

  • Resource Efficient Machine Learning: Making ML more efficient in terms of compute, energy consumption, training data and labels can impact their global adoption, while improving their overall environmental sustainability. Funded by two EU Horizon 2020 projects, we will explore these themes further in our new work. Results from our earlier work show some promising results in these directions. For instance, we introduced tensor networks for medical image classification tasks that require less than 10% of GPU resources when compared to traditional deep learning models. We have also put out a tool – Carbontracker – to predict and track carbon footprint of developing deep learning models, in an effort to encourage research into energy-efficient models which is used widely by the community.

  • AIChemy: Generative Modelling of nanoparticles: Obtaining structures from scattering data for nanomolecules turns out to be an expensive and difficult task. This ongoing collaboration with Prof. Kirsten Jensen’s group we are explorting generating valid nanomolecular structures from x-ray scattering data using deep generative models with promising early results.

  • Locomotion analysis: In this collabortion with Kiehn Lab, my role has been to analyze videos of mice running on treadmills in order to extract speed, cadence and coordination statistics. The results from analyzing some 1000 videos have accentuated the role of specific interneurons in the spinal cord that influence locomotion.

  • Sustainable neuroimaging: As the Chair of the SEA-SIG Neuroimaging Research Working Group, a team within SEA-SIG, I am leading an international team of 15 colleagues. This volunteer team is dedicated to the carbon footprinting of neuroimaging pipelines, and generation of best practice guidance.

  • Quantifying insect biodiversity: In this ongoing collaboration with FaunaPhotonics, we are attempting unsupervised clustering of insects based on optically recorded signals from field data. An initial attempt involved using a variational autoencoder that dynamically adjusts the tradeoff between reconstruction and regularisation terms.

  • Uncertainty Quantification: Medical image segmentation with well calibrated and meaningful uncertainty estimation can be more useful in clinical settings. For example, we present a probabilistic segmentation model that uses normalizing flows to improve the diversity of segmented samples.

Completed projects