Abstract: A method for training a control policy for controlling a technical system. The method includes training a neural network to implement a value function by: adapting the neural network for reducing a loss which, for a plurality of states and, for each state, for at least one action that has been previously carried out in the state, involves a deviation between a prediction for a cumulative reward and an estimation of the cumulative reward that is ascertained from a subsequent state that has been achieved by the action, and a reward that is obtained by the action. In the loss, for each action, the deviation for the action is weighted more strongly the greater the likelihood is that the action is selected by the control policy, in relation to the likelihood that the action is selected by a behavior control policy. The method also includes training the control policy.

Abstract: A method for controlling a robot device. The method includes: acquiring an image of an environment of the robot device; processing the image using a neural network, which outputs from the image a respective value image with pixel values for multiple pixels for at least one action parameter value, the pixel value for each pixel indicating an evaluation of an action specified by the action parameter value and the position of the pixel in the value image; selecting, from multiple actions, the particular action from among the multiple actions for which the pixel value of the pixel in the value image is at a maximum for the action parameter value at the position that specifies the action together with the action parameter value; and controlling the robot device to carry out the selected action.

Abstract: A method for training a conditional neural process for determining a position of an object from image data. The method includes: providing training data for training the conditional neural process, wherein the training data comprise labeled image data showing a particular object and labeled comparison image data regarding the particular object; and training the conditional neural process based on the provided training data, wherein the training of the conditional neural process comprises applying functional contrastive learning, and wherein the training of the conditional neural process comprises applying an end-to-end learning approach.

Abstract: A method for ascertaining a 6D pose of an object. The method includes the following steps: providing image data, wherein the image data include target image data showing the object and labeled comparison image data relating to the object, and ascertaining the 6D pose of the object based on the provided image data using a meta-learning algorithm.

Abstract: A robot device, a method for training a robot control model, and a method for controlling a robot device. The method for training includes: supplying an image showing object(s), to a first and second prediction model to produce a first and second pickup prediction that has, for each pixel of the image, a first and second pickup robot configuration vector with an assigned first and second success probability; supplying the first and second pickup prediction to a blending model of the robot control model to produce a third pickup prediction that has, for each pixel of the image: a third pickup robot configuration vector that is a weighted combination of the first and second pickup robot configuration vector, and a third success probability that is a weighted combination of the first and second success probability; and training the robot control model by adapting the first and second weighting factors.

Abstract: A method for controlling a robotic device. The method includes: obtaining an image, processing the image using a neural convolutional network, which generates an image in a feature space from the image, the image in the feature space, feeding the image in the feature space to a neural actor network, which generates an action parameter image, feeding the image in the feature space and the action parameter image to a neural critic network, which generates an assessment image, which defines for each pixel an assessment for the action defined by the set of action parameter values for that pixel, selecting, from multiple sets of action parameters of the action parameter image, that set of action parameter values having the highest assessment, and controlling the robot for carrying out an action according to the selected action parameter set.

Abstract: A method for training a control arrangement for a controlled system. The control arrangement includes a regulation device and an actuator that operates according to a control strategy. The method includes the generation of control actions by the regulation device, each control action being generated by detecting measured variables that indicate a state of the controlled system, ascertaining a correction term for the detected measured variables by the actuator according to the control strategy, adapting the detected measured variables using the correction term for the detected measured variables, and generating the control action by supplying the adapted measured variables to the regulation device as the actual value. The method further includes training the control strategy by reinforcement learning for maximizing the gain that is achieved by the generated control actions.