Marco Bagatella

Position / Function

Contact
Marco Bagatella
Department of Computer Science
Andreasstrasse 5, 8092 Zürich
Email: mbagatella[AT]ethz[DOT]ch
Website

Short CV

2022-present

PhD student

ETH-MPI CLS

2019-2021

MSc in Computer Science

ETH Zurich

Selected References

All Publications

Bagatella M, Olšák M, Rolínek M, et al. Planning from pixels in environments with combinatorially hard search spaces[J]. Advances in Neural Information Processing Systems, 2021, 34: 24707-24718.
Bagatella M, Christen S, Hilliges O. SFP: State-free priors for exploration in off-policy reinforcement learning[J]. arXiv preprint arXiv:2205.13528, 2022.
Bagatella M, Martius G. Goal-conditioned offline planning from curious exploration[J]. Advances in Neural Information Processing Systems, 2023, 36: 15358-15383.

Research Interests

Deep reinforcement learning, learning-based control

Teaching

Reinforcement Learning, 2023