Model-based Reinforcement Learning for Versatile Robots in the Real World

Informazioni relative al progetto

REAL-RL

ID dell’accordo di sovvenzione: 101045454

DOI

10.3030/101045454

Data della firma CE 30 Agosto 2022

Data di avvio 1 Gennaio 2023

Data di completamento 31 Dicembre 2027

Finanziato da

• European Research Council (ERC)

Costo totale

€ 1 998 500,00

Contributo UE

€ 1 998 500,00

Coordinato da

EBERHARD KARLS UNIVERSITAET TUEBINGEN
Germany

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Data Management Plan (1)

Data Management Plan

Pubblicazioni

Conference proceedings (14)

Efficient Learning of High Level Plans from Play

Autori: Armengol Urpi, N. and Bagatella, M. and Hilliges, O. and Martius, G. and Coros, S.
Pubblicato in: IEEE, 2023
Editore: IEEE
DOI: 10.1109/ICRA48891.2023.10161415

LPGD: A General Framework for Backpropagation through Embedded Optimization Layers

Autori: Paulus, A. and Martius, G. and Musil, V.
Pubblicato in: ICML International Conference on Machine Learning, 2024
Editore: PMLR

Learning with 3D rotations, a hitchhiker's guide to SO(3)

Autori: Geist, Andreas Reé and Frey, Jonas and Zhobro, Mikel and Levina, Anna and Martius, Georg
Pubblicato in: ICML International Conference on Machine Learning, 2024
Editore: PMLR

Bridging the Gap to Real-World Object-Centric Learning

Autori: Seitzer, Maximilian and Horn, Max and Zadaianchuk, Andrii and Zietlow, Dominik and Xiao, Tianjun and Simon-Gabriel, Carl-Johann and He, Tong and Zhang, Zheng and Schölkopf, Bernhard and Brox, Thomas and Locatello, Francesco
Pubblicato in: ICLR International Conference on Learning Representations, 2023
Editore: ICLR

Backpropagation through Combinatorial Algorithms: Identity with Projection Works

Autori: Sahoo, Subham and Paulus, Anselm and Vlastelica, Marin and Musil, Vit and Kuleshov, Volodymyr and Martius, Georg
Pubblicato in: ICLR International Conference on Learning Representations, 2023
Editore: ICLR

Goal-conditioned Offline Planning from Curious Exploration

Autori: Bagatella, Marco and Martius, Georg
Pubblicato in: NeurIPS/ Advances in Information Processing, 2023
Editore: Curran Associates

Learning Diverse Skills for Local Navigation under Multi-constraint Optimality

Autori: Jin Cheng, Marin Vlastelica, Pavel Kolev, Chenhao Li, Georg Martius
Pubblicato in: 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024
Editore: IEEE
DOI: 10.1109/ICRA57147.2024.10611629

Temporally Consistent Object-Centric Learning by Contrasting Slots

Autori: Manasyan, Anna and Seitzer, Maximilian and Radovic, Filip and Martius, Georg and Zadaianchuk, Andrii
Pubblicato in: CVPR: Conference on Computer Vision and Pattern Recognition, 2025
Editore: IEEE

Object-Centric Learning for Real-World Videos by Predicting Temporal Feature Similarities

Autori: Zadaianchuk, Andrii and Seitzer, Maximilian and Martius, Georg
Pubblicato in: NeurIPS/ Advances in Information Processing, 2023
Editore: Curran Associates

Pink Noise Is All You Need: Colored Noise Exploration in Deep Reinforcement Learning

Autori: Eberhard, Onno and Hollenstein, Jakob and Pinneri, Cristina and Martius, Georg
Pubblicato in: ICLR International Conference on Learning Representations, 2023
Editore: ICLR

Causal Action Influence Aware Counterfactual Data Augmentation

Autori: Urpi, N. A. and Bagatella, M. and Vlastelica, M. and Martius, G.
Pubblicato in: International Conference on Machine Learning (ICML), 2024
Editore: PMLR

On the Transfer of Object-Centric Representation Learning

Autori: Didolkar, A. R. and Zadaianchuk, A. and Goyal, A. and Mozer, M. C. and Bengio, Y. and Martius*, G. and Seitzer*, M.
Pubblicato in: ICLR International Conference on Learning Representations, 2025
Editore: PMLR

Benchmarking Offline Reinforcement Learning on Real-Robot Hardware

Autori: Gürtler, Nico and Blaes, Sebastian and Kolev, Pavel and Widmaier, Felix and Wüthrich, Manuel and Bauer, Stefan and Schölkopf, Bernhard and Martius, Georg
Pubblicato in: ICLR International Conference on Learning Representations, 2023
Editore: ICLR

Learning Hierarchical World Models with Adaptive Temporal Abstractions from Discrete Latent Dynamics

Autori: Gumbsch, Christian and Sajid, Noor and Martius, Georg and Butz, Martin V.
Pubblicato in: ICLR International Conference on Learning Representations, 2024
Editore: ICLR

Peer reviewed articles (1)

Offline vs. Online Learning in Model-based RL: Lessons for Data Collection Strategies

Autori: Chen, Jiaqi and Shi, Ji and Sancaktar, Cansu and Frey, Jonas and Martius, Georg
Pubblicato in: Reinforcement Learning Journal, 2025, ISSN 2996-8577
Editore: RLJ
DOI: 10.48550/arXiv.2509.05735