
Diego Hitzges
My research focuses on deep learning for structured problems in perception and decision-making. In my previous work, I developed efficient neural architectures for event-based depth estimation and combinatorial scheduling, with an emphasis on compact models, computational efficiency, and generalization. At the Zuse Institute, I aim to continue working on deep learning, potentially with a focus on efficient learning.
Contact
- office
- Room 3024 at ZIB
- email shown with JavaScript enabled
- homepage
- github.com/DiegoHitzges
- languages
- German, English, and Spanish
Curriculum vitae
- since 2026
- Researcher at ZIB
Publications and preprints
Conference contributions
- Hitzges, D., Ghosh, S., and Gallego, G. (2025). DERD-Net: Learning Depth from Event-based Ray Densities. Advances in Neural Information Processing Systems, 38, 1079–1103.
Associated with NeurIPS 2025.
DOI: 10.52202/085713-0035
[arXiv]
[BibTeX]
@inproceedings{2025_HitzgesGhoshGallego_EventRayDensities_2504-15863, year = {2025}, booktitle = {Advances in Neural Information Processing Systems}, series = {Advances in Neural Information Processing Systems}, volume = {38}, publisher = {Curran Associates}, pages = {1079--1103}, doi = {10.52202/085713-0035}, archiveprefix = {arXiv}, eprint = {2504.15863}, arxiv = {arXiv:2504.15863}, primaryclass = {cs.CV}, author = {Hitzges, Diego and Ghosh, Suman and Gallego, Guillermo}, title = {DERD-Net: Learning Depth from Event-based Ray Densities} } -
Associated with ICMLA 2025.
DOI: 10.1109/icmla66185.2025.00007
[arXiv]
[BibTeX]
@inproceedings{2025_HitzgesSagnol_DeepLearningUnrelatedMachines_2512-19527, year = {2025}, booktitle = {2025 International Conference on Machine Learning and Applications (ICMLA)}, publisher = {Institute of Electrical and Electronics Engineers}, pages = {1--8}, doi = {10.1109/icmla66185.2025.00007}, archiveprefix = {arXiv}, eprint = {2512.19527}, arxiv = {arXiv:2512.19527}, primaryclass = {cs.LG}, author = {Hitzges, Diego and Sagnol, Guillaume}, title = {Deep Learning for Unrelated-Machines Scheduling: Handling Variable Dimensions} }