Zobrazeno 1 - 10
of 244
pro vyhledávání: '"Martius Georg"'
Zero-shot imitation learning algorithms hold the promise of reproducing unseen behavior from as little as a single demonstration at test time. Existing practical approaches view the expert demonstration as a sequence of goals, enabling imitation with
Externí odkaz:
http://arxiv.org/abs/2410.08751
Pre-trained generalist policies are rapidly gaining relevance in robot learning due to their promise of fast adaptation to novel, in-domain tasks. This adaptation often relies on collecting new demonstrations for a specific task of interest and apply
Externí odkaz:
http://arxiv.org/abs/2410.05026
Identifying the physical properties of the surrounding environment is essential for robotic locomotion and navigation to deal with non-geometric hazards, such as slippery and deformable terrains. It would be of great benefit for robots to anticipate
Externí odkaz:
http://arxiv.org/abs/2408.16567
Linear temporal logic (LTL) is a powerful language for task specification in reinforcement learning, as it allows describing objectives beyond the expressivity of conventional discounted return formulations. Nonetheless, recent works have shown that
Externí odkaz:
http://arxiv.org/abs/2408.09495
Autor:
Didolkar, Aniket, Zadaianchuk, Andrii, Goyal, Anirudh, Mozer, Mike, Bengio, Yoshua, Martius, Georg, Seitzer, Maximilian
The goal of object-centric representation learning is to decompose visual scenes into a structured representation that isolates the entities. Recent successes have shown that object-centric representation learning can be scaled to real-world scenes b
Externí odkaz:
http://arxiv.org/abs/2408.09162
Embedding parameterized optimization problems as layers into machine learning architectures serves as a powerful inductive bias. Training such architectures with stochastic gradient descent requires care, as degenerate derivatives of the embedded opt
Externí odkaz:
http://arxiv.org/abs/2407.05920
Offline data are both valuable and practical resources for teaching robots complex behaviors. Ideally, learning agents should not be constrained by the scarcity of available demonstrations, but rather generalize beyond the training distribution. Howe
Externí odkaz:
http://arxiv.org/abs/2405.18917
Many settings in machine learning require the selection of a rotation representation. However, choosing a suitable representation from the many available options is challenging. This paper acts as a survey and guide through rotation representations.
Externí odkaz:
http://arxiv.org/abs/2404.11735
Autor:
Mattamala, Matías, Frey, Jonas, Libera, Piotr, Chebrolu, Nived, Martius, Georg, Cadena, Cesar, Hutter, Marco, Fallon, Maurice
Natural environments such as forests and grasslands are challenging for robotic navigation because of the false perception of rigid obstacles from high grass, twigs, or bushes. In this work, we present Wild Visual Navigation (WVN), an online self-sup
Externí odkaz:
http://arxiv.org/abs/2404.07110
Autor:
Schumacher, Pierre, Krause, Lorenz, Schneider, Jan, Büchler, Dieter, Martius, Georg, Haeufle, Daniel
Recent studies have demonstrated the immense potential of exploiting muscle actuator morphology for natural and robust movement -- in simulation. A validation on real robotic hardware is yet missing. In this study, we emulate muscle actuator properti
Externí odkaz:
http://arxiv.org/abs/2402.05371