Zobrazeno 1 - 10
of 116
pro vyhledávání: '"Doncieux, Stéphane"'
Autor:
Huber, Johann, Hélénon, François, Kappel, Mathilde, Chelly, Elie, Khoramshahi, Mahdi, Amar, Faïz Ben, Doncieux, Stéphane
Recent advances in AI have led to significant results in robotic learning, including natural language-conditioned planning and efficient optimization of controllers using generative models. However, the interaction data remains the bottleneck for gen
Externí odkaz:
http://arxiv.org/abs/2403.06173
Autor:
Sigaud, Olivier, Baldassarre, Gianluca, Colas, Cedric, Doncieux, Stephane, Duro, Richard, Oudeyer, Pierre-Yves, Perrin-Gilbert, Nicolas, Santucci, Vieri Giuliano
A lot of recent machine learning research papers have ``open-ended learning'' in their title. But very few of them attempt to define what they mean when using the term. Even worse, when looking more closely there seems to be no consensus on what dist
Externí odkaz:
http://arxiv.org/abs/2311.00344
Robotic grasping refers to making a robotic system pick an object by applying forces and torques on its surface. Many recent studies use data-driven approaches to address grasping, but the sparse reward nature of this task made the learning process c
Externí odkaz:
http://arxiv.org/abs/2310.04517
Despite recent advancements in AI for robotics, grasping remains a partially solved challenge, hindered by the lack of benchmarks and reproducibility constraints. This paper introduces a vision-based grasping framework that can easily be transferred
Externí odkaz:
http://arxiv.org/abs/2310.04349
Autor:
Salehi, Achkan, Doncieux, Stephane
Quality-Diversity is a branch of stochastic optimization that is often applied to problems from the Reinforcement Learning and control domains in order to construct repertoires of well-performing policies/skills that exhibit diversity with respect to
Externí odkaz:
http://arxiv.org/abs/2308.13278
Autor:
Salehi, Achkan, Doncieux, Stephane
Model-based RL/control have gained significant traction in robotics. Yet, these approaches often remain data-inefficient and lack the explainability of hand-engineered solutions. This makes them difficult to debug/integrate in safety-critical setting
Externí odkaz:
http://arxiv.org/abs/2303.01563
This paper studies the impact of the initial data gathering method on the subsequent learning of a dynamics model. Dynamics models approximate the true transition function of a given task, in order to perform policy search directly on the model rathe
Externí odkaz:
http://arxiv.org/abs/2210.11801
Robotics grasping refers to the task of making a robotic system pick an object by applying forces and torques on its surface. Despite the recent advances in data-driven approaches, grasping remains an unsolved problem. Most of the works on this task
Externí odkaz:
http://arxiv.org/abs/2210.07887
Publikováno v:
IEEE Transactions on Robotics; Print ISSN: 1552-3098; Online ISSN: 1941-0468
Model-based Reinforcement Learning and Control have demonstrated great potential in various sequential decision making problem domains, including in robotics settings. However, real-world robotics systems often present challenges that limit the appli
Externí odkaz:
http://arxiv.org/abs/2207.12062
Grasping a particular object may require a dedicated grasping movement that may also be specific to the robot end-effector. No generic and autonomous method does exist to generate these movements without making hypotheses on the robot or on the objec
Externí odkaz:
http://arxiv.org/abs/2205.08189