no code implementations • 2 Sep 2022 • Chang Rajani, Karol Arndt, David Blanco-Mulero, Kevin Sebastian Luck, Ville Kyrki
To this end we propose a co-imitation methodology for adapting behaviour and morphology by matching state distributions of the demonstrator.
1 code implementation • 29 Jun 2022 • Gabriele Tiboni, Karol Arndt, Giuseppe Averta, Ville Kyrki, Tatiana Tommasi
However, transferring the acquired knowledge to the real world can be challenging due to the reality gap.
no code implementations • 18 Apr 2022 • Ali Ghadirzadeh, Petra Poklukar, Karol Arndt, Chelsea Finn, Ville Kyrki, Danica Kragic, Mårten Björkman
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.
1 code implementation • 27 Jan 2022 • Rituraj Kaushik, Karol Arndt, Ville Kyrki
In this work, we introduce a novel learning algorithm called SafeAPT that leverages a diverse repertoire of policies evolved in the simulation and transfers the most promising safe policy to the real robot through episodic interaction.
1 code implementation • 20 Jan 2022 • Gabriele Tiboni, Karol Arndt, Ville Kyrki
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 difficult.
no code implementations • 25 May 2021 • Anton Mallasto, Karol Arndt, Markus Heinonen, Samuel Kaski, Ville Kyrki
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 with an affine transformation.
no code implementations • 12 Mar 2021 • Karol Arndt, Oliver Struckmeier, Ville Kyrki
Domain adaptation is a common problem in robotics, with applications such as transferring policies from simulation to real world and lifelong learning.
no code implementations • 16 Oct 2020 • Karol Arndt, Ali Ghadirzadeh, Murtaza Hazara, Ville Kyrki
Few-shot adaptation is a challenging problem in the context of simulation-to-real transfer in robotics, requiring safe and informative data collection.
no code implementations • 16 Sep 2019 • Karol Arndt, Murtaza Hazara, Ali Ghadirzadeh, Ville Kyrki
Modern reinforcement learning methods suffer from low sample efficiency and unsafe exploration, making it infeasible to train robotic policies entirely on real hardware.
2 code implementations • 10 Mar 2019 • Aleksi Hämäläinen, Karol Arndt, Ali Ghadirzadeh, Ville Kyrki
Training end-to-end deep robot policies requires a lot of domain-, task-, and hardware-specific data, which is often costly to provide.