Laura Manduchi
Ph.D. Candidate, Computer Science
ETH Zürich, Institute for Machine Learning
Medical Data Science group
Email: laura.manduchi at inf.ethz.ch



I am a PhD student in Computer Science at ETH Zürich under the supervision of Julia Vogt and Gunnar Rätsch. I am member of the Medical Data Science group and the ETH AI Centre. My research lies at the interplay between probabilistic modelling and deep learning, with a focus on representation learning, deep generative models, and clustering algorithms. I am particularly interested in incorporating domain knowledge in the form of constraints and probabilistic relations to obtain preferred representations of data that are robust to biases, with applications in medical imaging and X-ray astronomy.

I did my undergraduate studies in Information Engineering at the University of Padua, Italy, where I worked with Fabio Vandin on the Optimization of Fast Westfall-Young algorithm for mining significant patterns. I further obtained a M.Sc. in Data Science at ETH Zürich, where I acquired a strong background in Machine Learning. During my master's studies, I gained practical experience by working with Digited Galaxus, Switzerland, to implement subset selection algorithms using the Imitation Learning procedure. My Master’s thesis project under the supervision of Gunnar Rätsch was focused on the intersection between clustering and representation learning. After that, I did a research internship at the European Space Agency where I had the opportunity to apply state-of-the-art Machine Learning methods in astrophysics. In February 2020 I joined the Medical Data Science lab lead by Julia Vogt at ETH as a PhD student. During my PhD studies, I completed an internship at Microsoft Research Cambridge, working with Melanie F. Pradier on disentangled representantions of T-cell receptors repertoire. I am the co-leader of CSNOW, Computer Science Network of Women at ETH. I am supported by a PhD fellowship from the Swiss Data Science Center.


Publications

  1. Tree Variational Autoencoders.
    Laura Manduchi, Moritz Vandenhirtz, Alain Ryser, Julia E Vogt
    Spotlight NeurIPS, 2023.
    [paper][code]

  2. Deep Generative Clustering with Multimodal Diffusion Variational Autoencoders.
    Emanuele Palumbo, Laura Manduchi, Sonia Laguna, Daphne Chopard, Julia E Vogt
    ICLR, 2024.
    [paper]

  3. Learning Group Importance using the Differentiable Hypergeometric Distribution.
    Thomas M Sutter, Laura Manduchi, Alain Ryser, Julia E Vogt
    Spotlight ICRL, 2023.
    [paper]

  4. Interpretable Prediction of Pulmonary Hypertension in Newborns using Echocardiograms.
    L. Manduchi, H. Ragnarsdottir, H. Michel, F. Laumer, S. Wellmann, E. Ozkan, J. E. Vogt.
    GCPR 2022
    [paper]

  5. Signal Is Harder To Learn Than Bias: Debiasing with Focal Loss.
    Moritz Vandenhirtz, Laura Manduchi, Ricards Marcinkevics, Julia E Vogt
    Spotlight Domain Generalization Workshop, ICRL 2023.
    [paper]

  6. Weakly supervised inference of personalized heart meshes based on echocardiography videos.
    F. Laumer, M. Amrani, L. Manduchi, A. Beuret, A. Dubatovka, L. Rubi, C. Matter, J. M. Buhmann.
    Medical Image Analysis 2022.
    [paper]

  7. Anomaly Detection in Echocardiograms with Dynamic Variational Trajectory Models.
    Alain Ryser, Laura Manduchi, Fabian Laumer, Holger Michel, Sven Wellmann, Julia E Vogt
    Machine Learning for Healthcare (MLHC), 2022.
    [paper]

  8. A Deep Variational Approach to Clustering Survival Data.
    Laura Manduchi, Ricards Marcinkevics, Michela C. Massi, Thomas Weikert, Alexander Sauter, Verena Gotta, Timothy Müller, Flavio Vasella, Marian C. Neidert, Marc Pfister, Bram Stieltjes, Julia E. Vogt
    ICLR, 2022.
    Contributed talk. AI for Public Health Workshop, ICLR 2021.
    [paper][code]

  9. Deep Conditional Gaussian Mixture Model for Constrained Clustering.
    Laura Manduchi, Kieran Chin-Cheong, Holger Michel, Sven Wellmann, Julia E. Vogt.
    NeurIPS, 2021
    [paper][code]

  10. Deep Heart Beat: Latent trajectory learning of cardiac cycles using cardiac ultrasounds.
    Fabian Laumer, Gabriel Fringeli, Alina Dubatovka, Laura Manduchi, Joachim M. Buhmann
    Best Newcomer Award and Spotlight presentation. ML4H Workshop, NeurIPS 2020.
    [paper][code]

  11. T-DPSOM: an interpretable clustering method for unsupervised learning of patient health states
    Laura Manduchi, Matthias Hüser, Martin Faltys, Julia E. Vogt, Gunnar Rätsch, Vincent Fortuin.
    ACM CHIL 2021 and ML4H Workshop, NeurIPS 2019.
    [paper][code]

Invited Talks

Teaching

Service

Template Design by Aditya Grover.