Zobrazeno 1 - 10
of 24
pro vyhledávání: '"Arndt, Karol"'
Autor:
Bouniot, Quentin, Redko, Ievgen, Mallasto, Anton, Laclau, Charlotte, Arndt, Karol, Struckmeier, Oliver, Heinonen, Markus, Kyrki, Ville, Kaski, Samuel
In the last decade, we have witnessed the introduction of several novel deep neural network (DNN) architectures exhibiting ever-increasing performance across diverse tasks. Explaining the upward trend of their performance, however, remains difficult
Externí odkaz:
http://arxiv.org/abs/2310.11439
Autor:
Struckmeier, Oliver, Redko, Ievgen, Mallasto, Anton, Arndt, Karol, Heinonen, Markus, Kyrki, Ville
Optimal transport (OT) is a powerful geometric tool used to compare and align probability measures following the least effort principle. Despite its widespread use in machine learning (ML), OT problem still bears its computational burden, while at th
Externí odkaz:
http://arxiv.org/abs/2305.07500
The co-adaptation of robots has been a long-standing research endeavour with the goal of adapting both body and behaviour of a system for a given task, inspired by the natural evolution of animals. Co-adaptation has the potential to eliminate costly
Externí odkaz:
http://arxiv.org/abs/2209.01207
Physics simulators have shown great promise for conveniently learning reinforcement learning policies in safe, unconstrained environments. However, transferring the acquired knowledge to the real world can be challenging due to the reality gap. To th
Externí odkaz:
http://arxiv.org/abs/2206.14661
Autor:
Ghadirzadeh, Ali, Poklukar, Petra, Arndt, Karol, Finn, Chelsea, Kyrki, Ville, Kragic, Danica, Björkman, Mårten
We present a data-efficient framework for solving sequential decision-making problems which exploits the combination of reinforcement learning (RL) and latent variable generative models. The framework, called GenRL, trains deep policies by introducin
Externí odkaz:
http://arxiv.org/abs/2204.08573
The framework of Simulation-to-real learning, i.e, learning policies in simulation and transferring those policies to the real world is one of the most promising approaches towards data-efficient learning in robotics. However, due to the inevitable r
Externí odkaz:
http://arxiv.org/abs/2201.13248
In recent years, domain randomization over dynamics parameters has gained a lot of traction as a method for sim-to-real transfer of reinforcement learning policies in robotic manipulation; however, finding optimal randomization distributions can be d
Externí odkaz:
http://arxiv.org/abs/2201.08434
Sample-efficient domain adaptation is an open problem in robotics. In this paper, we present affine transport -- a variant of optimal transport, which models the mapping between state transition distributions between the source and target domains wit
Externí odkaz:
http://arxiv.org/abs/2105.11739
Domain adaptation is a common problem in robotics, with applications such as transferring policies from simulation to real world and lifelong learning. Performing such adaptation, however, requires informative data about the environment to be availab
Externí odkaz:
http://arxiv.org/abs/2103.07223
Few-shot adaptation is a challenging problem in the context of simulation-to-real transfer in robotics, requiring safe and informative data collection. In physical systems, additional challenge may be posed by domain noise, which is present in virtua
Externí odkaz:
http://arxiv.org/abs/2010.08397