Raghav{endra Selvan}

Raghav{endra Selvan}

Assistant Professor (TT) @ Machine Learning Section, University of Copenhagen.

Projects

Sustainability of AI

Material cost of developing and deploying complex ML models is growing considerably. In this line of research, we are focusing on some facets of the environmental sustainability of ML. Primarily, by focusing on energy consumption and carbon footprint.

  1. Raghavendra Selvan Bob Pepin, Christian Igel, Gabrielle Samuel, Erik B Dam. Equity through Access: A Case for Small-scale Deep Learning . Arxiv, 2024. (link) (pdf)
  2. Hallgrimur Thorsteinsson, Valdemar J Henriksen, Tong Chen, Raghavendra Selvan. Adversarial Fine-tuning of Compressed Neural Networks for Joint Improvement of Robustness and Efficiency . Arxiv, 2024. (link) (pdf)
  3. Tong Chen, Raghavendra Selvan. Is Adversarial Training with Compressed Datasets Effective? . Arxiv, 2024. (link) (pdf)
  4. Nicholas E Souter, Nikhil Bhagwat, Chris Racey, Reese Wilkinson, Niall W Duncan, Gabrielle Samuel, Loïc Lannelongue, Raghavendra Selvan, Charlotte L. Rae. Measuring and reducing the carbon footprint of fMRI preprocessing in fMRIPrep . OSF Preprint, 2024. (link) (pdf)
  5. Sebastian Eliassen, Raghavendra Selvan. Activation Compression of Graph Neural Networks using Block-wise Quantization with Improved Variance Minimization . International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2024. (link) (pdf)
  6. Pedram Bakhtiarifard, Christian Igel, Raghavendra Selvan. Energy Consumption-Aware Tabular Benchmarks for Neural Architecture Search . International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2024. (link) (pdf)
  7. Dustin Wright, Gabrielle Samuel, Christian Igel, Raghavendra Selvan. Efficiency is Not Enough: A Critical Perspective of Environmentally Sustainable AI . Arxiv, 2023. (link) (pdf)
  8. Nicholas E. Souter, Loïc Lannelongue, Gabrielle Samuel, Chris Racey, Lincoln J. Colling, Nikhil Bhagwat, Raghavendra Selvan, Charlotte L. Rae. Ten recommendations for reducing the carbon footprint of research computing in human neuroimaging . Imaging Neuroscience, 2023. (link) (pdf)
  9. Raghavendra Selvan, Julian Schön, Erik B Dam. Operating critical machine learning models in resource constrained regimes . Resource Efficient Medical Image Analysis Workshop at MICCAI2023, 2023. (link) (pdf)
  10. Nicklas Boserup, Raghavendra Selvan. Efficient Self-Supervision using Patch-based Contrastive Learning for Histopathology Image Segmentation . Northern Lights Deep Learning Conference, 2023. (link) (pdf)
  11. Raghavendra Selvan, Nikhil Bhagwat, Lasse F. Wolff Anthony, Benjamin Kanding, Erik B. Dam. Carbon Footprint of Selecting and Training Deep Learning Models for Medical Image Analysis . 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2022. (link) (pdf)
  12. Raghavendra Selvan. Carbon footprint driven deep learning model selection for medical imaging . 4th Conference on Medical Imaging with Deep Learning (MIDL), 2021. (link) (pdf)
  13. Lasse F Wolff Anthony, Benjamin Kanding, Raghavendra Selvan. Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models . ICML Workshop on Challenges in Deploying and monitoring Machine Learning Systems, 2020. (link) (pdf)

AI for Sciences

ML methods can accelerate research and open possibilities of asking novel questions in many scientific disciplines. In this line of research, several inter-disciplinary collaborations spanning a broad range of topics are being investigated. From an ML point of view, these do open interesting methods development.

  1. Roser Montanana-Rosell, Raghavendra Selvan, Pablo Hernández-Varas, Jan M. Kaminski, Dana B. Ahlmark, Ole Kiehn, Ilary Allodi. Spinal inhibitory neurons degenerate before motor neurons and excitatory neurons in a mouse model of ALS . (to appear in) Science Advances, 2024. (link) (pdf)
  2. Santiago Mora, Anna Stuckert, Rasmus von Huth Friis, Kimberly Pietersz, Gith Noes-Holt, Roser Montanana-Rosell, Haoyu Wang, Andreas Toft Sorensen, Raghavendra Selvan, Joost Verhaagen, Ilary Allodi. Stabilization of V1 interneuron-motor neuron connectivity ameliorates motor phenotype in a mouse model of ALS . (to appear in) Nature Communications, 2024. (link) (pdf)
  3. Ulrik Friis-Jensen, Frederik Lizak Johansen, Andy Sode Anker, Erik Bjørnager Dam, Kirsten Marie Ørnsbjerg Jensen, Raghavendra Selvan. CHILI: Chemically-Informed Large-scale Inorganic Nanomaterials Dataset for Advancing Graph Machine Learning . Arxiv, 2024. (link) (pdf)
  4. Frederik Lizak Johansen, Andy Sode Anker, Ulrik Friis-Jensen, Erik Bjørnager Dam, Kirsten Marie Ørnsbjerg Jensen, Raghavendra Selvan. A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation . Journal of Open Source Software (JOSS), 2024. (link) (pdf)
  5. Kenneth Thorø Martinsen, Kaj Sand-Jensen, Raghavendra Selvan. Predicting lake bathymetry from the topography of the surrounding terrain using deep learning . Limnology and Oceanography: Methods, 2023. (link) (pdf)
  6. Andy S. Anker, Kieth T. Butler, Raghavendra Selvan, Kirsten M.Ø. Jensen. Machine Learning for Analysis of Experimental Scattering and Spectroscopy Data in Materials Chemistry . Chemical Science, 2023. (link) (pdf)
  7. Haizea Goñi-Erro, Raghavendra Selvan, Roberto Leiras, Ole Kiehn. Pedunculopontine Chx10+ neurons control global motor arrest in mice . Nature Neuroscience, 2023. (link) (pdf)
  8. Emil T. S. Kjær, Andy S. Anker, Marcus N. Weng, Simon J. L. Billinge, Raghavendra Selvan, Kirsten M. Ø. Jensen. DeepStruc: Towards structure solution from pair distribution function data using deep generative models . Digital Discovery, 2023. (link) (pdf)
  9. Andy S. Anker, Emil T. S. Kjær, Mikkel Juelsholt, Troels Lindahl Christiansen, Susanne Linn Skjærvø, Mads Ry Vogel Jørgensen, Innokenty Kantor, Daniel Risskov Sørensen, Simon J. L. Billinge, Raghavendra Selvan, Kirsten M. Ø. Jensen. Extracting Structural Motifs from Pair Distribution Function Data of Nanostructures using Explainable Machine Learning . NPJ Computational Materials, 2022. (link) (pdf)
  10. Klas Rydhmer, Raghavendra Selvan. Dynamic β-VAEs for quantifying biodiversity by clustering optically recorded insect signals . Journal of Ecological Informatics, 2021. (link) (pdf)
  11. Ilary Allodi, Roser Montañana-Rosell, Raghavendra Selvan, Peter Low, Ole Kiehn. Locomotor deficits in a mouse model of ALS are paralleled by loss of V1-interneuron connections onto fast motor neurons . Nature Communications, 2021. (link) (pdf)
  12. Manh Cuong Ngo, Raghavendra Selvan, Outi Tervo, Mads Peter Heide-Jørgensen, Susanne Ditlevsen. Detection of foraging behavior from accelerometer data using U-Net type convolutional networks . Ecological Informatics, 2021. (link) (pdf)
  13. Andy Sode Anker, Emil TS Kjær, Erik B Dam, Simon JL Billinge, Kirsten MØ Jensen, Raghavendra Selvan. Characterising the Atomic Structure of Mono-Metallic Nanoparticles from X-Ray Scattering Data Using Conditional Generative Models . 16th International Workshop on Mining and Learning with Graphs, 2020. (link) (pdf)
  14. Abraham George Smith, Jens Petersen, Raghavendra Selvan, Camilla Ruø Rasmussen. Segmentation of roots in soil with U-Net . Journal of Plant Methods, 2019. (link) (pdf)

Bio-Medical Image Analysis

PhD training of RS was in medical image analysis. RS still holds keen interest in this domain and is active in investigating uncertainty quantification and use of deep latent generative models/ GNNs in this domain.

  1. Julian Schön, Raghavendra Selvan, Lotte Nygård, Ivan Richter Vogelius, Jens Petersen. Explicit Temporal Embedding in Deep Generative Latent Models for Longitudinal Medical Image Synthesis . Arxiv, 2023. (link) (pdf)
  2. Julian Schön, Raghavendra Selvan, Jens Petersen. Interpreting Latent Spaces of Generative Models for Medical Images using Unsupervised Methods . Deep Generative Models Workshop, MICCAI, 2022. (link) (pdf)
  3. Haozhe Luo, Yu Changdong, Raghavendra Selvan. Hybrid Ladder Transformers with Efficient Parallel-Cross Attention for Medical Image Segmentation . 5th International Conference on Medical Imaging with Deep Learning (MIDL), 2022. (link) (pdf)
  4. Justinas Antanavicius, Roberto Leiras Gonzalez, Raghavendra Selvan. Identifying partial mouse brain microscopy images from Allen reference atlas using a contrastively learned semantic space . International Workshop on Biomedical Image Registration, 2022. (link) (pdf)
  5. Raghavendra Selvan, Erik B Dam, Søren Alexander Flensborg, Jens Petersen. Patch-based medical image segmentation using Matrix Product State Tensor Networks . The Journal of Machine Learning for Biomedical Imaging (MELBA), 2022. (link) (pdf)
  6. Jan Mikolaj Kaminski, Ilary Allodi, Roser Montañana-Rosell, Raghavendra Selvan, Ole Kiehn. Deep ensemble model for segmenting microscopy images in the presence of limited labeled data . 4th Conference on Medical Imaging with Deep Learning (MIDL), 2021. (link) (pdf)
  7. Antonio Juarez, Raghavendra Selvan, Zaigham Saghir, HAWM Tiddens, Marleen de Bruijne. Automatic airway segmentation from Computed Tomography using robust and efficient 3-D convolutional neural networks . Scientific Reports, 2021. (link) (pdf)
  8. Raghavendra Selvan, Silas Ørting, Erik B Dam. Locally orderless tensor networks for classifying two- and three-dimensional medical images . The Journal of Machine Learning for Biomedical Imaging (MELBA), 2021. (link) (pdf)
  9. Raghavendra Selvan, Erik B Dam, Jens Petersen. Segmenting two-dimensional structures with strided tensor networks . 27th international conference on Information Processing in Medical Imaging (IPMI), 2021. (link) (pdf)
  10. Espen Jimenez-Solem, Tonny S Petersen, Casper Hansen, Christian Hansen, Christina Lioma, Christian Igel, Wouter Boomsma, Oswin Krause, Stephan Lorenzen, Raghavendra Selvan, Janne Petersen, Martin Erik Nyeland, Mikkel Zöllner Ankarfeldt, Gert Mehl Virenfeldt, Matilde Winther-Jensen, Allan Linneberg, Mostafa Mehdipour Ghazi, Nicki Detlefsen, Andreas David Lauritzen, Abraham George Smith, Marleen de Bruijne, Bulat Ibragimov, Jens Petersen, Martin Lillholm, Jon Middleton, Stine Hasling Mogensen, Hans-Christian Thorsen-Meyer, Anders Perner, Marie Helleberg, Benjamin Skov Kaas-Hansen, Mikkel Bonde, Alexander Bonde, Akshay Pai, Mads Nielsen, Martin Sillesen. Developing and validating COVID-19 adverse outcome risk prediction models from a bi-national European cohort of 5594 patients . Scientific Reports, 2021. (link) (pdf)
  11. Raghavendra Selvan, Silas Ørting, Erik B Dam. Multi-layered tensor networks for image classification . First Workshop on Quantum Tensor Networks in Machine Learning. In conjunction with 34th NeurIPS, 2020. (link) (pdf)
  12. Raghavendra Selvan, Frederik Faye, Jon Middleton, Akshay Pai. Uncertainty quantification in medical image segmentation with Normalizing Flows . 11th International Workshop on Machine Learning in Medical Imaging, 2020. (link) (pdf)
  13. Raghavendra Selvan, Erik B Dam, Nicki Skafte Detlefsen, Sofus Rischel, Kaining Sheng, Mads Nielsen, Akshay Pai. Lung Segmentation from Chest X-rays using Variational Data Imputation . ICML Workshop on The Art of Learning with Missing Values (Artemiss-2020), 2020. (link) (pdf)
  14. Raghavendra Selvan, Erik B Dam. Tensor Networks for Medical Image Classification . Proceedings of International Conference on Medical Imaging with Deep Learning (MIDL). Winner of Runner-up Best Paper Award., 2020. (link) (pdf)
  15. Raghavendra Selvan, Thomas Kipf, Max Welling, Antonio GU Juarez, Jesper H. Pedersen, Jens Petersen, Marleen de Bruijne. Graph Refinement based Airway Extraction using Mean-Field Networks and Graph Neural Networks . Medical Image Analysis, 2020. (link) (pdf)
  16. Jens Petersen, Andres M Arias-Lorza, Raghavendra Selvan, Daniel Bos, Aad van der Lugt, Jesper H Pedersen, Mads Nielsen, Marleen de Bruijne. Increasing Accuracy of Optimal Surfaces Using Min-Marginal Energies . IEEE transactions on medical imaging, 2019. (link) (pdf)
  17. Raghavendra Selvan, Jens Petersen, Jesper H. Pedersen, Marleen de Bruijne. Extracting tree-structures in CT data by tracking multiple statistically ranked hypotheses . International Journal of Medical Physics Research and Practice, 2019. (link) (pdf)
  18. Antonio Juarez, Raghavendra Selvan, Zaigham Saghir, Marleen de Bruijne. A joint 3D UNet-Graph Neural Network-based method for Airway Segmentation from chest CTs . the proceedings of International Workshop on Machine Learning in Medical Imaging, 2019. (link) (pdf)
  19. Raghavendra Selvan, Max Welling, Jesper H. Pedersen, Jens Petersen, Marleen de Bruijne. Mean field network based graph refinement with application to airway tree extraction . 21st Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2018), Grenada, Spain, 2018. (link) (pdf)
  20. Raghavendra Selvan. Extraction of Airways from Volumetric Data . University of Copenhagen (PhD Thesis), 2018. (link) (pdf)
  21. Raghavendra Selvan, Thomas Kipf, Max Welling, Jesper H. Pedersen, Jens Petersen, Marleen de Bruijne. Extraction of Airways using Graph Neural Networks . 1st Conference on Medical Imaging with Deep Learning (MIDL 2018), Amsterdam., 2018. (link) (pdf)
  22. Raghavendra Selvan, Jens Petersen, Jesper H. Pedersen, Marleen de Bruijne. Extraction of airways with probabilistic state-space models and Bayesian smoothing . Graphs in Biomedical Image Analysis, Computational Anatomy and Imaging Genetics, 2017. (link) (pdf)
  23. Raghavendra Selvan, Jens Petersen, Jesper H. Pedersen, Marleen de Bruijne. Extraction of airway trees using multiple hypothesis tracking and template matching . Sixth International Workshop on Pulmonary Image Analysis. MICCAI, 2016. (link) (pdf)