Signal Processing Systems
Dept. of Microelectronics

MSc C. Meo

PhD student
Signal Processing Systems (SPS), Department of Microelectronics

PhD thesis (Feb 2026): World Models: Foundations, Applications, and Limitations
Promotor: Justin Dauwels, Geert Leus

Expertise: Machine learning

Themes: Machine learning for image and video understanding

Biography

Cristian Meo was a PhD student at SPS under supervision of Justin Dauwels and Geert Leus, and received his PhD degree (cum laude) in Feb 2026. In his research, he investigated how different inductive biases affect neural networks generalization and reasoning capabilities. 

In particular, he studied how inter and intra class factors of variations can be disentangled within the modular networks framework. In other words, looking at both object centric and attribute level disentanglement.  In order to do that, he studies how a network's capacity is linked to disentanglement using information-theoretic constructs.

After his PhD, Cristian created a startup LatentWorlds AI.

EE4685 Bayesian machine learning

Mathematical foundation for machine learning algorithms, presented from a statistical (Bayesian) and optimization point of view.

  1. World Models: Foundations, Applications, and Limitations
    Cristian Meo;
    PhD thesis, TU Delft, Feb. 2026.
    document

  2. Precipitation Nowcasting Using Physics Informed Discriminator Generative Models
    Junzhe Yin; Cristian Meo; Ankush Roy; Zeineh Bou Cher; Mircea Lica; Yanbo Wang; Ruben Imhoff; Remko Uijlenhoet; Justin Dauwels;
    In 32nd European Signal Processing Conference, EUSIPCO 2024, Lyon, France, August 26-30, 2024,
    IEEE, pp. 967--971, 2024.
    document

  3. alphaTC-VAE: On the relationship between Disentanglement and Diversity
    Cristian Meo; Louis Mahon; Anirudh Goyal; Justin Dauwels;
    In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024,
    OpenReview.net, 2024.
    document

  4. Nowcasting of Extreme Precipitation Using Deep Generative Models
    Haoran Bi; Maksym Kyryliuk; Zhiyi Wang; Cristian Meo; Yanbo Wang; Ruben Imhoff; Remko Uijlenhoet; Justin Dauwels;
    In IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023, Rhodes Island, Greece, June 4-10, 2023,
    IEEE, pp. 1--5, 2023. DOI: 10.1109/ICASSP49357.2023.10094988
    document

BibTeX support

Last updated: 13 Feb 2026

Cristian Meo

Alumnus