Abstract: A method for training a machine learning model. The method includes: ascertaining, for each of a plurality of training data elements, a gradient of a target function, wherein the gradient comprises a component for each of a plurality of parameters of the machine learning model; generating an overall gradient by averaging the ascertained gradients component-wise by summing, for each component, the values of the ascertained gradients for this component and by dividing a resulting sum for the component by the number of ascertained gradients for which the component is above a specified threshold value; and adjusting the machine learning model in a direction given by the overall gradient.

Abstract: A method for generating a state model describing a controllable system. The method includes: providing at least one part of the state model; selecting an action from a set of actions starting from the second state of the components; simulating further states of the components by a successive application of an action from the set of actions to the components in each case, an individual reward being determined for each of the applications of an action to the components; optimizing the at least one part of the state model based on the determined rewards, the optimizing of the at least one part of the state model taking place based on a variance reduction method and a maximum of the determined rewards; and adding the selected action and the second state to the at least one part of the state model.

Abstract: A method for training a control strategy with the aid of reinforcement learning. The method includes carrying out passes, in each pass, an action that is to be carried out being selected for each state of a sequence of states of an agent, for at least some of the states the particular action being selected by specifying a planning horizon that predefines a number of states, ascertaining multiple sequences of states, reachable from the particular state, using the predefined number of states, by applying an answer set programming solver to an answer set programming program which models the relationship between actions and the successor states that are reached by the actions, selecting the sequence that delivers the maximum return, and selecting an action as the action for the particular state via which the first state of the selected sequence may be reached, starting from the particular state.

Abstract: A method for an industrial system. The method includes: ascertaining a representation of the industrial system, the ascertainment of the representation including: selecting a first state of the representation, selecting, based on the first state, at least one machining order from a plurality of machining orders as a function of the first state of the representation and as a function of at least one previously ascertained recommendation, and ascertaining a second state as a subsequent state of the first state via a simulation of the second state as a function of the at least one selected machining order and as a function of the first state; and ascertaining a manufacturing schedule for the industrial system as a function of the ascertained representation.

Abstract: A method for scheduling a set of jobs for a plurality of machines. Each job is defined by at least one feature which characterizes a processing time of the job. If any of the machines is free, a job from of the set of jobs is selected to be carrying out by said machine and scheduled for said machine. The job is selected as follows: a Graph Neural Network receives as input the set of jobs and a current state of at least the machine which is free, the Graph Neural Network outputs a reward for the set of jobs if launched on the machines, which states are inputted into the Graph Neuronal Network, and the job for the free machine is selected depending on the Graph Neural Network output.