Abstract: A method is for behavior planning for an at least partially automated ego vehicle in a traffic scene, in which a current state of the traffic scene is captured and mapped to an initial state in a latent space. The method includes starting from the initial state, a tree structure of a plurality of state sequences in the latent space is generated by generating at least one subsequent state of the tree structure for n successive time steps, each starting from a starting state that the at least one subsequent state of the tree structure is generated. The method also includes determining, for the starting state, a multimodal behavior policy for the ego vehicle and a starting value of the starting state, and selecting a selected behavior option from the multimodal behavior policy and determining a reward for selecting the selecting behavior option in the starting state.

Abstract: A method for training a driving policy for autonomous vehicles. The method includes: pre-training the driving policy in a first stage using one of the following techniques: i) Reinforcement Learning (RL) with a hand-crafted reward function, or ii) behavior cloning from expert demonstrations; storing intermediate results from the first stage, including: i) rollouts generated by the policy, and/or ii) model weights of the pre-trained policy; applying Inverse Reinforcement Learning (IRL), comprising: i) learning a reward model from expert demonstrations, and/or ii) fine-tuning the pre-trained policy from the first stage with RL under the learned reward model; characterized in that the stored intermediate results of the first stage are utilized during the secondary stage, comprising: i) using the stored rollouts to augment the expert demonstration dataset, and ii) reusing the stored model weights to initialize the policy and the reward model.

Abstract: A method for training a control device for a robotic apparatus. The method includes: ascertaining a training dataset with a set of training trajectories; ascertaining an imitation loss between the training trajectories and trajectories supplied by the control device, wherein the imitation loss for each training trajectory has a single loss to a particular trajectory supplied by the control device; wherein a weighting factor for the single loss is ascertained depending on the trajectory supplied by the control device and the single loss is weighted in the imitation loss by the weighting factor, wherein the weighting factor is piecewise constant with respect to the trainable parameters of the control model, and training the control device to reduce the imitation loss.

Abstract: An amber emitting nitride ceramic phosphor may be directly sintered under pressure to improve the resulting compositional gradient. As a result mechanical thinning of the ceramic is less necessary. The ceramic also comes out more chemically stable, allowing for the example the application of a dichroic layer which can improve color point in the final light emitting device. The inventive process is cheaper than conventional production methods and may results in a phosphor with better quantum efficiency.

Abstract: A device and a method for training at least one model or for generating a first agent, in particular for an interaction between the first agent and a second agent. A description of a behavior of the first agent, in particular in the interaction between the first agent and the second agent, is mapped onto a first representation using a first model; the first representation is mapped onto a second representation by means of a second model; the second representation is mapped onto an output variable for influencing the behavior of the first agent using a third model. The description is specified in natural language, in text form or audio form, or in formal language or in digital graphic form, wherein the behavior of the first agent, in particular in the interaction between the first agent and the second agent, is specified depending on the output variable.

Abstract: A computer-implemented method for performing Learning from Demonstrations, particularly Imitation Learning, based on data associated with a first domain, particularly a source domain. The method includes: determining first data characterizing a demonstrator of the first domain, wherein particularly the first data characterizes sensor data of the demonstrator and/or sensor data of at least one spectator observing the demonstrator, determining first knowledge from the first domain based on the first data, transferring at least a part of the first knowledge to a second domain, particularly a target domain.

Abstract: A device and method for controlling a hardware agent in a control situation having a plurality of hardware agents. The method includes ascertaining of a potential function by a first neural network; ascertaining of a control scenario for a control situation from a plurality of possible control scenarios by a second neural network; ascertaining a common action sequence for the plurality of hardware agents by seeking an optimum of the ascertained potential function over the possible common action sequences of the ascertained control scenario; and controlling at least one of the plurality of hardware agents in accordance with the ascertained common action sequence.

Abstract: A computer-implemented method for training a machine learning system. The machine learning system is configured to determine a control signal characterizing an action to be executed by a technical system. The method includes obtaining a safe action to be executed by the technical system including: obtaining a state signal; determining, by a parametrized policy module of the machine learning system, a distribution of potentially unsafe actions that could be executed by the technical system; sampling a potentially unsafe action from the distribution; obtaining, by a safety module of the machine learning system, the safe action. The method further includes determining a loss value based on the state signal and the safe action; and training the machine learning system by updating parameters of the policy module according to a gradient of the loss value with respect to the parameters.

Abstract: A device and method for controlling a hardware agent in a control situation having a plurality of hardware agents. The method includes ascertaining of a potential function by a first neural network; ascertaining of a control scenario for a control situation from a plurality of possible control scenarios by a second neural network; ascertaining a common action sequence for the plurality of hardware agents by seeking an optimum of the ascertained potential function over the possible common action sequences of the ascertained control scenario; and controlling at least one of the plurality of hardware agents in accordance with the ascertained common action sequence.

Abstract: A computer-implemented method for performing Learning from Demonstrations, particularly Imitation Learning, based on data associated with a first domain, particularly a source domain. The method includes: determining first data characterizing a demonstrator of the first domain, wherein particularly the first data characterizes sensor data of the demonstrator and/or sensor data of at least one spectator observing the demonstrator, determining first knowledge from the first domain based on the first data, transferring at least a part of the first knowledge to a second domain, particularly a target domain.