Abstract: A training data set obtaining method includes determining, based on degrees of contribution of transmission conditions to an optimization objective of a neural network, data volumes of training data under the transmission conditions; and obtaining, based on the data volumes of the training data under the transmission conditions, the training data under the transmission conditions, to form a training data set for training the neural network. The degrees of contribution of the transmission conditions to the optimization objective of the neural network represent degrees of impact of the transmission conditions on a value of the optimization objective of the neural network.

Abstract: Disclosed herein are methods for identifying the structure of patterns and anomalies in flow of events from the cyber-physical system or information system. In one aspect, an exemplary method comprises, using at least one connector, getting event data, generating at least one episode consisting of a sequence of events, and transferring the generated episodes to an event processor; and using the event processor, process episodes using a neurosemantic network, wherein the processing includes recognizing events and patterns previously learned by the neurosemantic network, training the neurosemantic network, identifying a structure of patterns by mapping to the patterns of neurons on a hierarchy of layers of the neurosemantic network, attributing events and patterns corresponding to neurons of the neurosemantic network to an anomaly depending on a number of activations of the corresponding neuron, and storing the state of the neurosemantic network.

Abstract: A method for knowledge representation and deduction of service logic includes: generating, based on a knowledge representation model, a semantic graph corresponding to conceptual-layer service logic, where the semantic graph includes one or more types of nodes and edges for connecting the one or more types of nodes, and the nodes include at least a node of a variable type; generating, based on the semantic graph and a physical table to which a service object is mapped, an instance graph, where the instance graph includes the nodes and edges in the semantic graph; generating executable code based on a service logic relationship between the nodes in the instance graph; and determining, based on the executable code and a data instance corresponding to a node whose in-degree is 0 in the instance graph, a data instance corresponding to each node in the instance graph.

Abstract: An embodiment for self-development of resources is provided. The embodiment may include receiving data relating to an activity and a first robotic device assigned to perform the activity. The embodiment may also include creating a knowledge corpus of a second set of one or more robotic devices capable of performing the activity. The embodiment may further include executing a digital twin simulation of a digital twin model of the first robotic device performing the activity. The embodiment may also include in response to determining the first robotic device is unable to complete the activity without incident, identifying within the second set of one or more robotic devices a most comparable robotic device to the first robotic device. The embodiment may further include predicting a modification of the first robotic device. The embodiment may also include attaching one or more resources printed by a 3D printer to the first robotic device.

Abstract: One or more computer processors improve knowledge base question answering (KBQA) model convergence and prediction performance by generalizing the KBQA model based on transfer learning.

Abstract: The invention discloses a vehicle fault reasoning method based on a knowledge graph, includes: constructing a knowledge graph of a vehicle fault; obtaining a question statement (QS) from user; performing question classification on the QS by TextCNN to obtain a classification result; marking a training QS by using an NER data marking manner and training an entity extraction model based on a sequence marking result; generating a vehicle fault class by a decision tree model for extracting the QS and searching for an answer in the knowledge graph; and performing question template matching based on the classification result and the answer and substituting the answer into the question template to obtain an answer statement. A question answering system constructed by the method includes word processing, question classification, question template matching, and answer generation, which helps the user judging a fault problem of vehicle, and searching for related knowledge about vehicle fault.

Abstract: The application relates to a method for developing the agitation system of a scale-up polymerization vessel. A simulated prediction model is obtained by use of a small polymerization vessel and by integrating Taguchi experimental design method with artificial intelligence (AI) neural network. Accordingly, vessel parameters for the agitation system of a scale-up polymerization vessel can be rapidly and accurately predicted based on simulation qualities thereof, further facilitating a construction of the agitation system of a scale-up polymerization vessel.

Abstract: A method for generating descriptions of classification results, comprising classifying original inputs in a data set using a classifier. For one or more of the plurality of original inputs, apply an adjustment to one or more features of the original input to produce a respective adjusted input, and classify the respective adjusted input using the classifier. For one or more of the adjusted inputs, record a corresponding original input into a training set, and record information associated with the adjusted input into the training set. The method further comprises training a model based on the training set, such that the model is configured to receive an original model input and identify a hypothetical adjustment that would produce an adjusted model input that would be assigned an adjusted model classification with a different class than a class assigned to the original model input.

Abstract: In variants, the method for determining a sample recommendation can include: determining characteristic values for a sample, determining target characteristic values, determining a similarity score for the sample based on the characteristic values for the sample and the target characteristic values, training a prediction model, and determining a sample recommendation.

Abstract: A model-assisted system for predicting survivability of a patient may include at least one processor. The processor may be programmed to access a database storing a medical record for the patient. The medical record may include at least one of structured and unstructured information relative to the patient and may lack a structured patient ECOG score. The processor may be further programmed to analyze at least one of the structured and unstructured information relative to the patient; based on the analysis, and in the absence of a structured ECOG score, generate a performance status prediction for the patient; and provide an output indicative of the predicted performance status. The analysis of at least one of the structured and unstructured information and the generation of the predicted performance status may be performed by at least one of a trained machine learning model or a natural language processing algorithm.

Abstract: A logical expression processing apparatus includes at least one processor, the at least one processor carrying out: an acquisition process of acquiring background knowledge information and query information; a processing execution process of processing the logical formulas of at least one of the background knowledge information and the query information that have been acquired in the acquisition process, and acquiring processed background knowledge information and processed query information; an inference execution process of executing logical inference using, as inputs, the processed background knowledge information and the processed query information; and an evaluation process of outputting at least one of the processed background knowledge information and the processed query information, depending on whether or not an execution result of the logical inference satisfies a predetermined requirement.

Abstract: A semiconductor metrology system including a spectrum acquisition tool for collecting, using a first measurement protocol, baseline scatterometric spectra on first semiconductor wafer targets, and for various sources of spectral variability, variability sets of scatterometric spectra on second semiconductor wafer targets, the variability sets embodying the spectral variability, a reference metrology tool for collecting, using a second measurement protocol, parameter values of the first semiconductor wafer targets, and a training unit for training, using the collected spectra and values, a prediction model using machine learning and minimizing an associated loss function incorporating spectral variability terms, the prediction model for predicting values for production semiconductor wafer targets based on their spectra.

Abstract: An apparatus for physiologically informed gestational inquiries is disclosed. The apparatus includes at least a processor and memory communicatively connected to the at least a processor. The memory instructs the processor to receive a biological extraction from the user. The memory instructs the processor to receive a gestational inquiry from the user. The memory instructs the processor to separate the gestational inquiry from a description of the gestational inquiry. The memory instructs the processor to determine a gestational target as a function of the gestational inquiry and the biological extraction. The memory instructs the processor to generate a gestational report as a function of the gestational target.

Abstract: A dataset generation system has a camera classification circuitry configured to classify plural cameras into plural groups, an input device for a user to set a selecting criterion of a captured image, a first captured image collection circuitry configured to collect captured images which are captured by at least one camera in each group classified by the camera classification circuitry, and which meet the selecting criterion set by a user using the input device, an inference circuitry configured to perform inference processing on each of the captured images collected by the first captured image collection circuitry and a dataset evaluation circuitry configured to evaluate whether or not a dataset consisting of the captured images collected by the first captured image collection circuitry is suitable for a training dataset of a neural network model for a predetermined inference process, based on the result of the inference processing by the inference circuitry.

Abstract: Embodiments described herein provide a processing apparatus comprising compute circuitry to generate neural network data for a convolutional neural network (CNN) and write the neural network data to a memory buffer. The compute circuitry additionally includes a direct memory access (DMA) controller including a hardware codec having encode circuitry and a decode circuitry. The DMA controller reads the neural network data from the memory buffer, encode the neural network data via the encode circuit, writes encoded neural network data to a memory device coupled with the processing apparatus, writes metadata for the encoded neural network data to the memory device coupled with the processing apparatus, and decodes encoded neural network data via the decode circuit in response to a request from the compute circuitry.

Abstract: A method for generating one or more quantum circuits to be evaluated by one or more QPUs is disclosed. The method includes obtaining a problem to be solved using one or more QPUs and analyzing the problem for reductions in complexity to obtain a reduced problem. One or more quantum algorithms are generated to implement a potential solution to the reduced problem, and the quantum algorithms are translated into one or more quantum circuits. The method further includes transmitting the one or more quantum circuits to the one or more QPUs.

Abstract: A Hierarchical Ensembles of Autonomous Decision Systems (HEADS) system with a recursive ensemble weighting update is proposed. The system is built on fuzzy logic leading to an understandable and tractable logic design that leverages subject matter experts to design system operations. The hierarchical structure enables multi-layered logic for granular control and decisions incorporating inferred information. The control output from each ensemble is a mixture from independently trained fuzzy systems processed through a gating network. The gating network weights are updated recursively. Each expert uses a subset of the input space to minimize per-expert complexity and support ensemble robustness under uncertain or evolving state realizations and operating environments. Finally, autonomy based on fuzzy systems offers the potential for increased human comprehension of an agent's status and decision logic.

Abstract: Methods and apparatus for performing quantum annealing using a quantum system. In one aspect, a method includes controlling the quantum system such that a total Hamiltonian characterizing the quantum system evolves from an initial quantum Hamiltonian to a problem quantum Hamiltonian, wherein controlling the quantum system comprises applying an inhomogeneous driving field to the quantum system to drive the quantum system across a quantum phase transition.

Abstract: Aspects of the subject disclosure may include, for example, providing, by a global node, a model to a group of nodes of a network, the model being associated with determining a configuration of the network for qubits; receiving, by the global node from one or more nodes of the group of nodes, updated model parameters, wherein the updated model parameters are generated by each of the one or more nodes via training of a local model utilizing the model and local data accessible to the particular node of the one or more nodes resulting in local nodes; generating, by the global node, an updated model based on the updated model parameters, the updated model being associated with determining the configuration of the network for the qubits; and providing, by the global node, the updated model to the group of nodes of the network, wherein the updated model facilitates managing distribution and usage of entangled qubit storage in devices of the network as reserve hybrid quantum-classical network capacity for bandwidth

Abstract: There is provided a quantum computing method and device. The method comprises generating a pair of communication qubits having a quantum entanglement; transporting each of the pair of communication qubits to two different cores, respectively; executing an operation between a data qubit positioned at each of the two different cores and each of the pair of the communication qubits transported to each of the two different cores; measuring a quantum state of the pair of communication qubits; and executing an operation on the data qubit positioned at each of the two different cores based on a measurement result of the quantum state of the pair of communication qubits.

Abstract: One embodiment provides a method and a system for diagnosing a digital circuit. During operation, the system can obtain a design of the digital circuit, generate a design of a diagnostic circuit by augmenting the design of the digital circuit based on a number of fault-emulating subcircuits, and convert the design of the diagnostic circuit to a design of a quantum oracle circuit. The system can further construct a quantum diagnostic circuit based on the design of the quantum oracle circuit and observe states of the quantum diagnostic circuit to determine probability distributions of one or more faults in the digital circuit.

Abstract: A device comprises a data quantum bit, a first quantum bit coupler, a second quantum bit coupler, and an auxiliary quantum bit. The first quantum bit coupler is coupled to the data quantum bit. The second quantum bit coupler is coupled to the first quantum bit coupler. The auxiliary quantum bit is coupled to the second quantum bit coupler. The first quantum bit coupler is configured to operate in a state to suppress interaction between the data quantum bit and the auxiliary quantum bit. The first quantum bit coupler and the second quantum bit coupler are each configured to operate in a respective state to enable interaction between the data quantum bit and the auxiliary quantum bit and entangle a state of the data quantum bit with a state of the auxiliary quantum bit.

Abstract: A controller of a quantum system causes a geometric phase gate to be performed on two or more qubits by causing two or more qubits of the quantum system to experience a (near field and/or non-radiating) magnetic field gradient; and, responsive to determining that a gate time period has elapsed since the two or more qubits of the quantum system started to experience the magnetic field gradient, causing the two or more qubits to no longer experience the magnetic field gradient. The entanglement of the two or more qubits corresponding to the performance of the gate is enacted, mediated, and/or caused by the magnetic field gradient.

Abstract: Quantum information processing involves entangling large numbers of qubits, which can be realized as defect centers in a solid-state host. The qubits can be implemented as individual unit cells, each with its own control electronics, that are arrayed in a cryostat. Free-space control and pump beams address the qubit unit cells through a cryostat window. The qubit unit cells emit light in response to these control and pump beams and microwave pulses applied by the control electronics. The emitted light propagates through free space to a mode mixer, which interferes the optical modes from adjacent qubit unit cells for heralded Bell measurements. The qubit unit cells are small (e.g., 10 ?m square), so they can be tiled in arrays of up to millions, addressed by free-space optics with micron-scale spot sizes. The processing overhead for this architecture remains relatively constant, even with large numbers of qubits, enabling scalable large-scale quantum information processing.

Abstract: A device may receive input data that includes demographic data, power demand data, power source data, power route data, technology data, industry data, and problem data associated with a geographic location, and may identify a section of the geographic location from the demographic data. The device may identify power sources of the section, and may estimate power generation and power demand for the section. The device may determine whether the power demand is greater than the power generation for the section. The device may utilize a quantum computer and a power optimizer model with the input data associated with the section to determine optimized power insights for the section based on determining that the power demand is greater than the power generation for the section, and may perform actions based on the optimized power insights for the section.

Abstract: The present disclosure relates to a passive outlier-tolerant and active outlier-tolerant learning method with abnormal samples and a system thereof. The method includes the following steps: (1) judging the distribution of abnormal samples in a train set, and calling a corresponding outlier-tolerant algorithm based on the distribution to carry out feature learning on the sample feature quantity in the train set to obtain an outlier-tolerant learning result; (2) training an initial learning model by using the outlier-tolerant learning result to obtain an intermediate learning model; (3) processing the test set by a preset outlier-correction algorithm to obtain an outlier-corrected test set; (4) testing the intermediate learning model using the outlier-corrected test set to obtain a final learning model. The scheme of the present disclosure can improve the passive outlier tolerance and active outlier tolerance of a machine learning process to abnormal samples.

Abstract: A method for training a control system model includes introducing a simulated fault into a software simulation of a physical system and generating emulated sensor data based on the simulated fault, where the emulated sensor data emulates output from one or more sensors of the physical system. The method further includes obtaining output data from a test control system provided with the emulated sensor data, where the test control system emulates a control system of the physical system and tagging the output data with the simulated fault to create training data. The method further includes utilizing the training data to train the control system model, where the control system model is a machine learning model for use with the control system of the physical system during operation of the physical system.

Abstract: Platform behavior may be adapted via machine-learning-based entity lifecycle monitoring. A web crawler agent collects data comprising an entity identifier token. A machine learning model is trained to determine, based at least in part on the entity identifier token, whether a corresponding entity is associated with the computing platform (e.g., whether the corresponding entity is a platform subscriber entity for the computing platform). Based on the output of the machine learning model applied to the entity identifier token (in some embodiments, in combination with other relevant data parsed from the collected data), an indication of an entity lifecycle status and a confidence value therefor are determined. Based on the entity lifecycle status and the confidence value, a listener is bound to the platform subscriber entity. The listener monitors activity of the platform subscriber entity with respect to the platform and identifies platform action to take in response.

Abstract: Systems and methods described herein are adapted to predict the intent of a network operator when it is determined that configuration changes (or config changes) are made to a network device. A method, according to one implementation, includes the step of collecting raw data related to a Network Element (NE). The method also includes the step of pre-processing the raw data to obtain a config change associated with the NE. Also, the method includes the step of applying the config change as input to a ML model to predict a user intent representing a desired network outcome.

Abstract: A validation method determining device includes a data acquiring unit to acquire training data to be used for training of a predictor and test data to be used for measurement of prediction accuracy of the predictor, a validation method recommending unit to recommend a validation method for extracting the training data and validation data corresponding to a prediction target using data type information indicating data types of the training data, the test data, and column data constituting both the data, and a validation method determining unit to determine a validation method to be used for estimation of prediction performance of the predictor from a recommended validation method.

Abstract: An information processing apparatus includes a holding unit configured to hold first input data having a plurality of fields, for use in machine learning, a generation unit configured to generate second input data having a plurality of fields, for use in the machine learning, based on the first input data, and an adjustment unit configured to adjust values stored in the plurality of fields in the second input data.

Abstract: A method for training at least one machine learning algorithm used to output specifications for interventions in the control system of a motor vehicle during specific driving maneuvers is disclosed. The method includes (i) during an operation of the motor vehicle, determining whether at least one specific driving maneuver is being performed; (ii) in the event that a specific driving maneuver is being performed, detecting interventions by a driver of the motor vehicle in the control system of the motor vehicle during the performance of the driving maneuver; and (iii) training the at least one machine learning algorithm based on the specific driving maneuver and the detected interventions by the driver of the motor vehicle in the control system of the motor vehicle.

Abstract: Disclosed is a method of integrally optimizing different types of parameters that require setting during a machine learning process. The disclosed method of integrally optimizing the parameters includes performing training on a machine learning model by selecting sensor parameters and machine learning model hyperparameters until a predetermined termination condition is satisfied; and determining, among the selected sensor parameters and machine learning model hyperparameters, an optimized sensor parameter and optimized machine learning model hyperparameter that minimize a loss value for the machine learning model, wherein the performing of the training on the machine learning model includes selecting the sensor parameters and machine learning model hyperparameters that satisfy a predetermined optimization range, and performing training on the machine learning model based on sensor data provided from a sensor by the selected sensor parameters and the machine learning model hyperparameters.

Abstract: An iterative method is for determining training data for a primary model to solve a primary recognition task. The iterative method includes a) providing at least one labeled training sample, b) training the primary model with the at least one labeled training sample, c) providing at least one labeled test sample, and d) evaluating a recognition performance of the primary model using the labeled test sample on the primary recognition task. The iterative method further includes, depending on a result of the evaluating the recognition performance, either (i) re-performing parts a), b), c), and d) of the iterative method, or (ii) ending the iterative method.

Abstract: The present disclosure provides systems and methods for end-to-end machine learning. A method of the present disclosure may comprise one or more operations of data ingestion, data preparation, feature storage, model building, and productionizing by the model. The methods and systems of the present disclosure may use an Automated Machine Learning (AutoML) algorithm and eXplainable Artificial Intelligence (XAI).

Abstract: Various embodiments of methods and systems, including computer programs encoded on computer storage media described herein are directed to real-time situation determination based on distributed event data. According to various embodiments, the system receives event data from one or more computing devices. The system provides a machine learning model configured to use a plurality of interconnected check-point evaluators to evaluate the received event data and determine an occurrence of a situation. The system evaluates event values, via one or more check-point evaluator of the plurality of interconnected check-point evaluators, whether the event values meet criteria for one or more situation indicators. Based on the evaluation of the event values the system determines the occurrence of the situation.

Abstract: Systems and methods for providing insights about a machine learning model are provided. The method includes, using training data to train the machine learning model to learn patterns to determine whether data associated with an event provides an indication that the event belongs to a certain class from among a plurality of classes, evaluating one or more features of the machine learning model to produce a data set pairing observed scores S and a set of predictive input variables Vi, and constructing at least one data-driven estimator based on an explanatory statistic, the estimator being represented in a computationally efficient form and packaged with the machine learning model and utilized to provide a definition of explainability for a score generated by the machine learning model.

Abstract: A system for providing ticket based authorized entry of a mobile access device includes a memory, a transceiver and a control module. The memory stores an access code of a ticket of an event to be accessed by the mobile access device. The mobile access device also stores the access code. The control module: determines an amount of time since the ticket was purchased or an amount of time since the access code of the ticket was last updated; and updates the access code stored in the memory and signals the mobile access device via the transceiver to update the access code stored at the mobile access device in response to the amount of time since the ticket was purchased exceeding a first predetermined amount of time or the amount of time since the access code of the ticket was last updated exceeding a second predetermined amount of time.

Abstract: A method, a computer-readable medium, and an apparatus for transport service are provided. The apparatus may receive a plurality of transport service transactions associated with a point of interest entity. For each transport service transaction, the apparatus may determine a transport service location at which the transport service transaction is executed. The apparatus may cluster the transport service locations determined for the plurality of transport service transactions. The apparatus may determine one or more candidate transport service locations for the point of interest entity based on the clustering. The apparatus may provide the one or more candidate transport service locations to a client or a service provider associated with a transport service transaction that is to be executed at the point of interest entity.

Abstract: According to one embodiment, a method, computer system, and computer program product for determining acoustics in a workplace is provided. The present invention may include performing digital twin simulation of sound generation and at least one machine in the workplace; calculating one or more resultant sound fields in the workplace; identifying sound cancellation potential of a worker's sound cancellation protection; and determining if the worker is wearing proper sound cancellation protection based on the one or more resultant sound fields in the workplace and the sound cancellation potential of the worker's sound cancellation protection.

Abstract: Computer-implemented method for generating control data configured to be used or usable in an agricultural equipment for treating a field, comprising the following steps: (S10) providing crop data, wherein the crop data comprise information about an agricultural crop species providing crop data grown or sown or planned to be grown or sown in a field; (S20) providing field data, wherein the field data comprise information about the field; (S30)—optionally—providing historic treatment data, (S40)—optionally—providing environmental data, (S50) at least based on the crop data and on the field data, initiating and/or performing data processing in at least one database and/or database system, (S60) determining the organism-specific risk based on the result of the data processing, (S70) providing and/or determining the organism-specific threshold, (S80) determining, based on the organism-specific risk and the organism-specific threshold and based on the data processing in at least one treatment-related database, at

Abstract: Systems and methods for detecting collaboration sessions and other workloads in a heterogeneous computing platform are described. In some embodiments, an Information Handling System (IHS) may include a heterogeneous computing platform having a plurality of devices and a memory coupled to the heterogeneous computing platform, where the memory includes a plurality of sets of firmware instructions, where each set of firmware instructions, upon execution by a respective device, enables the respective device to provide a corresponding firmware service, and where at least one of the plurality of devices operates as an orchestrator configured to collect telemetry from the plurality of devices and detect a collaboration session based, at least in part, upon the telemetry.

Abstract: A system may determine an initial safety prediction for a driver or rider associated with a shared mobility service. Then the system may receive, from various sensors, sensor data collected during operation of a vehicle. The system may determine a subset of the sensor data related to one or more shared mobility statuses, then determine, based on the subset of the sensor data, a safety score for the driver and/or rider. The system may further perform operations based on the safety score and/or the initial safety prediction.

Abstract: The present invention provides a processing apparatus 1 including: a situation information generation unit 10 that, based on a result of image analysis, generates situation information indicating a situation of each of a plurality of processes executed by a plurality of workers; a delayed process extraction unit 20 that extracts a delayed process the progress of which is delayed from target progress from among a plurality of the processes, based on the situation information; an estimation unit 30 that estimates a cause of a delay in the progress of the delayed process, based on the situation information; and an advice information generation unit 40 that, based on the progress of a process other than the delayed process and a cause of a delay in the progress of the delayed process, generates advice information about a method for assigning personnel to a plurality of the processes.

Abstract: An information processing apparatus includes a controller that acquires information about an article to be loaded onto a vehicle, acquires a characteristic of a person who can affect the article based on the information, and executes a concurrent riding determination to determine existence of concurrent riding on the vehicle by the article and a user having the characteristic.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating account prioritization scores for various accounts based on experience data and organization data and providing visual representations of the account prioritization scores via a graphical user interface. In particular, in one or more embodiments, the disclosed systems organize experience data and organization data to utilize in generating the account prioritization score. Further, in some embodiments, the disclosed systems generate and provide an explanation of a key factor for the account prioritization score. Thus, the disclosed systems provide an accurate and efficient graphical user interface to enable quick comprehension of even very large datasets corresponding to various accounts.

Abstract: A data management system that enables appropriate management of data in offices is provided. A data management system including one or more server devices provided in each of a plurality of offices, in which: when receiving a request for data of another office from a client terminal, a setting unit sets information indicating acquisition of data of the other office in setting information; and an acquisition unit acquires the data of the other office from the server device in the other office according to the setting information in which a change is detected by a monitoring unit.

Abstract: Example methods and systems are directed to a managed inventory. A database may store information regarding items owned by a user. The information regarding an item may include a quantity owned and one or more triggering events. Based on the occurrence of a triggering event, an order for the item may be placed without user intervention. Data to the database may be provided by one or more sensors. Triggering events may be defined in terms of sensor data. The triggering event may be defined by a user or through machine learning. The order may be placed using a predetermined modality or a dynamically-determined modality based on one or more criteria, such as price, shipping speed, and the urgency of the order.

Abstract: One or more computing devices, systems, and/or methods for constructing and utilizing a search structure for identifying tasks to assign to entities are provided. The search structure is constructed to partition tasks and represent time constraints for performing the tasks and distance constraints corresponding to locations of the tasks. A search of the search structure is performed to identify a set of nearest tasks with respect to a target task. Tasks within the set of nearest tasks may be assigned to an entity to perform.

Abstract: Aspects concern a method for controlling vehicles to perform transport tasks comprising supplying information about vehicles and information about transport tasks to a graph neural network by associating each vehicle with a vehicle graph node and each transport task with a transport task graph node, processing the vehicle and the transport graph by the neural network, wherein the neural network determines a feature for each graph node, determining, for each pair of a transport graph node and vehicle graph node, a weight representing a similarity between the features determined for the transport graph node and the vehicle graph node, selecting an assignment between the transport graph nodes and the vehicle graph nodes from a set of possible assignments, wherein the selected assignment maximizes the sum of the weights of the pairs and controlling each vehicle according to the selected assignment.