Marcel Hirt

Research Fellow

Nanyang Technological University

Marcel is currently a PostDoc at NTU Singapore, working with Victoria Leong and Juan-Pablo Ortega. He has been doing his PhD with Petros Dellaportas and Alex Beskos at UCL.
His research interests include
• Machine Learning: Generative Models, Multi-Modal Learning, Time Series Models.
• Computational Statistics: Variational Inference, Sequential Monte Carlo, Markov Chain Monte Carlo.
• Bayesian and Probabilistic Modeling: Epigenetics, Finance and Neuroscience.

Interests
  • Machine Learning
  • Computational Statistics
  • Probabilistic Modeling
Education

  • PhD in Statistics, 2021

    University College London

  • MSc in Financial Mathematics and Actuarial Sciences, 2014

    Technical University of Munich

  • BSc in Mathematics, 2011

    Technical University of Munich

Recent Publications

Marcel Hirt, Domenico Campolo, Victoria Leong, Juan-Pablo Ortega (2023). Learning multi-modal generative models with permutation-invariant encoders and tighter variational bounds. In arXiv.

PDF Cite

Marcel Hirt, Vasileios Kreouzis, Petros Dellaportas (2023). Learning variational autoencoders via MCMC speed measures. In arXiv.

PDF Cite Code

Giovanni Ballarin, Petros Dellaportas, Lyudmila Grigoryeva, Marcel Hirt, Sophie van Huellen, Juan-Pablo Ortega (2023). Reservoir computing for macroeconomic forecasting with mixed frequency data. In arXiv.

PDF Cite Code

Marcel Hirt, Axel Finke, Alexandros Beskos, Petros Dellaportas, Stephan Beck, Ismail Moghul, Simone Ecker (2022). A Bayesian framework for genome-wide inference of DNA methylation levels. In arXiv.

PDF Cite

Marcel Hirt, Michalis Titsias, Petros Dellaportas (2021). Entropy-based adaptive Hamiltonian Monte Carlo. In NeurIPS 2021.

PDF Cite Code

See all publications

Contact