The Experts below are selected from a list of 10014 Experts worldwide ranked by ideXlab platform

Dominik Henrich - One of the best experts on this subject based on the ideXlab platform.

  • guiding robots to Predefined Goal positions with multi modal feedback
    2018
    Co-Authors: Michael Riedl, Dominik Henrich
    Abstract:

    Manually guiding a robot to a Predefined Goal position is nearly impossible, because the user needs to align the tool center point of the robot precisely with the Predefined position. In this paper we present an approach to enable guiding a robot to a Goal position with the help of multi-modal feedback in form of haptic feedback by increasing the stiffness of the robot arm and visual feedback by showing the Goal position and the degree to which this position is reached in a graphic simulation window. First, we describe our approach that utilizes multi-modal feedback, then we explain the user study used to evaluate our approach and in the end, an overview on the results of the evaluation is given.

Oudeyer Pierre-yves - One of the best experts on this subject based on the ideXlab platform.

  • GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
    2021
    Co-Authors: Kovač Grgur, Laversanne-finot Adrien, Oudeyer Pierre-yves
    Abstract:

    Autonomous agents using novelty based Goal exploration are often efficient in environments that require exploration. However, they get attracted to various forms of distracting unlearnable regions. To solve this problem, absolute learning progress (ALP) has been used in reinforcement learning agents with Predefined Goal features and access to expert knowledge. This work extends those concepts to unsupervised image-based Goal exploration. We present the GRIMGEP framework: it provides a learned robust Goal sampling prior that can be used on top of current state-of-the-art novelty seeking Goal exploration approaches, enabling them to ignore noisy distracting regions while searching for novelty in the learnable regions. It clusters the Goal space and estimates ALP for each cluster. These ALP estimates can then be used to detect the distracting regions, and build a prior that enables further Goal sampling mechanisms to ignore them. We construct an image based environment with distractors, on which we show that wrapping current state-of-the-art Goal exploration algorithms with our framework allows them to concentrate on interesting regions of the environment and drastically improve performances. The source code is available at https://sites.google.com/view/grimgep

  • GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
    2021
    Co-Authors: Kovač Grgur, Laversanne-finot Adrien, Oudeyer Pierre-yves
    Abstract:

    Designing agents, capable of learning autonomously a wide range of skills is critical in order to increase the scope of reinforcement learning. It will both increase the diversity of learned skills and reduce the burden of manually designing reward functions for each skill. Self-supervised agents, setting their own Goals, and trying to maximize the diversity of those Goals have shown great promise towards this end. However, a currently known limitation of agents trying to maximize the diversity of sampled Goals is that they tend to get attracted to noise or more generally to parts of the environments that cannot be controlled (distractors). When agents have access to Predefined Goal features or expert knowledge, absolute Learning Progress (ALP) provides a way to distinguish between regions that can be controlled and those that cannot. However, those methods often fall short when the agents are only provided with raw sensory inputs such as images. In this work we extend those concepts to unsupervised image-based Goal exploration. We propose a framework that allows agents to autonomously identify and ignore noisy distracting regions while searching for novelty in the learnable regions to both improve overall performance and avoid catastrophic forgetting. Our framework can be combined with any state-of-the-art novelty seeking Goal exploration approaches. We construct a rich 3D image based environment with distractors. Experiments on this environment show that agents using our framework successfully identify interesting regions of the environment, resulting in drastically improved performances. The source code is available at https://sites.google.com/view/grimgep

Michael Riedl - One of the best experts on this subject based on the ideXlab platform.

  • guiding robots to Predefined Goal positions with multi modal feedback
    2018
    Co-Authors: Michael Riedl, Dominik Henrich
    Abstract:

    Manually guiding a robot to a Predefined Goal position is nearly impossible, because the user needs to align the tool center point of the robot precisely with the Predefined position. In this paper we present an approach to enable guiding a robot to a Goal position with the help of multi-modal feedback in form of haptic feedback by increasing the stiffness of the robot arm and visual feedback by showing the Goal position and the degree to which this position is reached in a graphic simulation window. First, we describe our approach that utilizes multi-modal feedback, then we explain the user study used to evaluate our approach and in the end, an overview on the results of the evaluation is given.

Kovač Grgur - One of the best experts on this subject based on the ideXlab platform.

  • GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
    2021
    Co-Authors: Kovač Grgur, Laversanne-finot Adrien, Oudeyer Pierre-yves
    Abstract:

    Autonomous agents using novelty based Goal exploration are often efficient in environments that require exploration. However, they get attracted to various forms of distracting unlearnable regions. To solve this problem, absolute learning progress (ALP) has been used in reinforcement learning agents with Predefined Goal features and access to expert knowledge. This work extends those concepts to unsupervised image-based Goal exploration. We present the GRIMGEP framework: it provides a learned robust Goal sampling prior that can be used on top of current state-of-the-art novelty seeking Goal exploration approaches, enabling them to ignore noisy distracting regions while searching for novelty in the learnable regions. It clusters the Goal space and estimates ALP for each cluster. These ALP estimates can then be used to detect the distracting regions, and build a prior that enables further Goal sampling mechanisms to ignore them. We construct an image based environment with distractors, on which we show that wrapping current state-of-the-art Goal exploration algorithms with our framework allows them to concentrate on interesting regions of the environment and drastically improve performances. The source code is available at https://sites.google.com/view/grimgep

  • GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
    2021
    Co-Authors: Kovač Grgur, Laversanne-finot Adrien, Oudeyer Pierre-yves
    Abstract:

    Designing agents, capable of learning autonomously a wide range of skills is critical in order to increase the scope of reinforcement learning. It will both increase the diversity of learned skills and reduce the burden of manually designing reward functions for each skill. Self-supervised agents, setting their own Goals, and trying to maximize the diversity of those Goals have shown great promise towards this end. However, a currently known limitation of agents trying to maximize the diversity of sampled Goals is that they tend to get attracted to noise or more generally to parts of the environments that cannot be controlled (distractors). When agents have access to Predefined Goal features or expert knowledge, absolute Learning Progress (ALP) provides a way to distinguish between regions that can be controlled and those that cannot. However, those methods often fall short when the agents are only provided with raw sensory inputs such as images. In this work we extend those concepts to unsupervised image-based Goal exploration. We propose a framework that allows agents to autonomously identify and ignore noisy distracting regions while searching for novelty in the learnable regions to both improve overall performance and avoid catastrophic forgetting. Our framework can be combined with any state-of-the-art novelty seeking Goal exploration approaches. We construct a rich 3D image based environment with distractors. Experiments on this environment show that agents using our framework successfully identify interesting regions of the environment, resulting in drastically improved performances. The source code is available at https://sites.google.com/view/grimgep

Laversanne-finot Adrien - One of the best experts on this subject based on the ideXlab platform.

  • GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
    2021
    Co-Authors: Kovač Grgur, Laversanne-finot Adrien, Oudeyer Pierre-yves
    Abstract:

    Autonomous agents using novelty based Goal exploration are often efficient in environments that require exploration. However, they get attracted to various forms of distracting unlearnable regions. To solve this problem, absolute learning progress (ALP) has been used in reinforcement learning agents with Predefined Goal features and access to expert knowledge. This work extends those concepts to unsupervised image-based Goal exploration. We present the GRIMGEP framework: it provides a learned robust Goal sampling prior that can be used on top of current state-of-the-art novelty seeking Goal exploration approaches, enabling them to ignore noisy distracting regions while searching for novelty in the learnable regions. It clusters the Goal space and estimates ALP for each cluster. These ALP estimates can then be used to detect the distracting regions, and build a prior that enables further Goal sampling mechanisms to ignore them. We construct an image based environment with distractors, on which we show that wrapping current state-of-the-art Goal exploration algorithms with our framework allows them to concentrate on interesting regions of the environment and drastically improve performances. The source code is available at https://sites.google.com/view/grimgep

  • GRIMGEP: Learning Progress for Robust Goal Sampling in Visual Deep Reinforcement Learning
    2021
    Co-Authors: Kovač Grgur, Laversanne-finot Adrien, Oudeyer Pierre-yves
    Abstract:

    Designing agents, capable of learning autonomously a wide range of skills is critical in order to increase the scope of reinforcement learning. It will both increase the diversity of learned skills and reduce the burden of manually designing reward functions for each skill. Self-supervised agents, setting their own Goals, and trying to maximize the diversity of those Goals have shown great promise towards this end. However, a currently known limitation of agents trying to maximize the diversity of sampled Goals is that they tend to get attracted to noise or more generally to parts of the environments that cannot be controlled (distractors). When agents have access to Predefined Goal features or expert knowledge, absolute Learning Progress (ALP) provides a way to distinguish between regions that can be controlled and those that cannot. However, those methods often fall short when the agents are only provided with raw sensory inputs such as images. In this work we extend those concepts to unsupervised image-based Goal exploration. We propose a framework that allows agents to autonomously identify and ignore noisy distracting regions while searching for novelty in the learnable regions to both improve overall performance and avoid catastrophic forgetting. Our framework can be combined with any state-of-the-art novelty seeking Goal exploration approaches. We construct a rich 3D image based environment with distractors. Experiments on this environment show that agents using our framework successfully identify interesting regions of the environment, resulting in drastically improved performances. The source code is available at https://sites.google.com/view/grimgep