Marco Fraccaro

       Ph.D. student

       Cognitive Systems, DTU Compute

       Technical University of Denmark (DTU)


       e-mail: marfra at dtu dot dk







Main research interests

·       Machine Learning

·       Bayesian Inference: stochastic and deterministic approximation techniques such as Markov Chain Monte Carlo methods, Sequential Monte Carlo methods, Expectation Propagation and Variational Inference.

·       Deep Learning, and more specifically using deep neural networks to parametrize flexible Variational distributions.

·       Recommender systems. For a detailed description of the project see this link.


If you want to know a bit more about me, you can take a look at my CV or LinkedIn profile.





Semi-Supervised Generation with Cluster-aware Generative Models (arXiv, pdf)

Lars Maaløe, Marco Fraccaro, Ole Winther

arXiv:1704.00637, 2017


Sequential Neural Models with Stochastic Layers (arXiv, pdf, supplementary, nips, code) [NIPS 2016 oral presentation]
Marco Fraccaro, Søren Kaae Sønderby, Ulrich Paquet, Ole Winther

Advances in Neural Information Processing Systems 29, NIPS 2016

arXiv:1605.07571, 2016


An Adaptive Resample-Move Algorithm for Estimating Normalizing Constants (arXiv, pdf)
Marco Fraccaro, Ulrich Paquet, Ole Winther

arXiv:1604.01972, 2016


An efficient implementation of Riemannian Manifold Hamiltionian Monte Carlo for Gaussian Process Models (pdf)
Ulrich Paquet, Marco Fraccaro

Technical report, 2016


Indexable Probabilistic Matrix Factorization for Maximum Inner Product Search (pdf)
Marco Fraccaro, Ulrich Paquet and Ole Winther
The Thirtieth AAAI Conference on Artificial Intelligence, 2016


Perturbation Theory for Variational Inference (pdf)
Manfred Opper, Marco Fraccaro, Ulrich Paquet, Alex Susemihl, and Ole Winther
NIPS 2015 Workshop on Advances in Approximate Bayesian Inferences, 2015


Learning to Index (pdf)

Marco Fraccaro

Master thesis, Technical University of Denmark, 2014


Palm area detection for reliable hand gesture recognition (pdf)
Giulio Marin, Marco Fraccaro, Mauro Donadeo, Fabio Dominio, and Pietro Zanuttigh
Proceedings of IEEE MMSP 2013, Pula, Italy, 2013