Fujitsu Patente

A detection program causes a computer to execute a process includes acquiring an input image from a camera acquiring a first detection result by inputting the input image to a first detection model that performs object detection acquiring, a plurality of detection results by inputting the input image to a plurality of detection models trained based on different training data sets with a parameter of the first detection model used as an initial value, and generating a detection result in which the first detection result and the detection results are combined.

An information processing apparatus includes a processor configured to generate a set of a plurality of routes for a plurality of moving bodies that depart from a specific node and return to the specific node via a plurality of nodes, the plurality of routes representing an order in which the plurality of moving bodies visit the nodes generate, by using the generated set of routes, a route selection problem with constraints on a shipment volume of a node serving as a shipment source of packages solve the generated route selection problem and output a result of solving the route selection problem.

A first wiring made of a superconductor and a first electrode made of a superconductor connected to the first wiring are formed on a substrate. A resist mask that covers at least the first wiring is formed. A surface of the first electrode using the resist mask is etched. An insulator film on the surface of the first electrode using the resist mask is formed. A second electrode made of a superconductor on the surface of the insulator film using the resist mask is formed.

A computer-implemented method comprising: generating, based on a plurality of candidate output tokens generated by a large language model, LLM, a plurality of candidate token embeddings, respectively and generating, based on at least one string describing prohibited content, at least one content embedding; comparing the candidate token embeddings with the at least one content embedding to determine for each candidate output token a similarity score indicating similarity between the candidate output token and the prohibited/unwanted content; for each candidate output token, generating, based on the similarity score and based on a selection probability assigned by the LLM to the candidate output token, an output score; and selecting, based on the output scores, a candidate output token for output by the LLM.

A computer implemented method for detecting computer generated images comprising: loading an input image, inputting the input image and a representation describing the input image into a denoising model for denoising the input image using the representation, generating a denoised image embedding from the denoised image, generating an input image embedding from the input image, comparing a difference between the input image embedding and the denoised image embedding with a similarity decision threshold to determine whether the input image is a real image or a computer generated image.

A computer-implemented method for traffic simulation, the method comprising: executing a preliminary agent-based simulation model to generate a preliminary simulation prediction of traffic in a geographic area; accepting input of traffic data for the geographic area; performing data assimilation on the preliminary simulation prediction and the traffic data to generate assimilated traffic data.

An information processing program that causes a computer to perform a process including: classifying a plurality of items to be determined into one or more categories based on a plurality of appearance images of the plurality of items, the plurality of appearance images being stored in a storage area (41a) in association with corresponding prices of the plurality of items; calculating, for each of the one or more categories into which the items are classified, a statistical value of prices of a plurality of items in the category, based on the plurality of prices stored in the storage area (41a); changing a pattern for the classifying and performing the classifying and the calculating so that one or more of the statistical values satisfy a given condition; and outputting category information (41b) indicating a category of each of the plurality of items when the given condition is satisfied.

In an embodiment, workflow for timeseries forecasting may be performed based on automated machine learning. Sensor data for measurement parameter is received from plurality of sensors installed in built environment and the received sensor data is stored in table of relational database. Cut-off record associated with previous training checkpoint is determined of the forecasting model for the measurement parameter. Records including new records are determined for which respective timestamps occur after the measurement timestamp of cut-off record. Size of the determined records are compared with threshold size and training dataset is prepared. The forecasting model is trained on the training dataset based on the comparison.

A processing unit determines, from a plurality of solutions to a combinatorial optimization problem stored in a storage unit, a plurality of higher-ranked solutions based on evaluation function values of the plurality of solutions, determines, for each of the plurality of higher-ranked solutions, a selection probability such that higher-ranked solutions having evaluation function values closer to that of the highest-ranked solution among the plurality of higher-ranked solutions are more likely to be selected, selects a first solution from the plurality of higher-ranked solutions according to the selection probabilities, selects a second solution different from the first solution, from the plurality of solutions, generates a third solution with a path relinking method using the selected first solution and second solution, and performs a solution search on the combinatorial optimization problem using the third solution as an initial solution.

A storing unit stores area information indicating positions of a plurality of image areas set in a captured image of a front region of a self-checkout terminal including a scanner. A processing unit recognizes a product from the captured image. The processing unit detects a fraudulent action related to a scanning operation for causing the scanner to scan product information attached to the product, based on a movement path of the recognized product in the plurality of image areas, and a residence time of the product in a first image area that is closest to the scanner among the plurality of image areas.

An answer generation program causes a computer to execute a process including acquiring a question input to a first agent that generates information based on input information, the question being related to a video for monitoring a specific task, identifying a specific agent that has a function of either video recognition for the specific task or domain knowledge of the specific task, from among a plurality of second agents capable of cooperating with the first agent, based on the acquired question, and causing the first agent to output, as an answer to the question, an answer result based on generation information that is generated by the identified specific agent in accordance with an instruction from the first agent.

According to an aspect of at least one embodiment, a method to improve a user interface may include obtaining transcript data include one or more words from a transcription of speech in the audio data. The transcript data may be generated by automated speech recognition technology from the audio data. The transcript data and criteria may be provided to a large language model configured to analyze the transcript data based on the criteria to select a word from the transcript data. A definition of the selected word may also be generated by the large language model. The selected word and the definition of the selected word may be obtained. The audio data may be broadcasted by a device and the selected word and the definition of the selected word may be presented on a display of the device with the broadcasting of the audio data.

A qubit device (100) includes a first qubit substrate (10a) having a first qubit (30a) formed on a first surface (12a), a second qubit substrate (10b) having a second qubit (30b) formed on a second surface (12b), a connection substrate (40) disposed to face the first surface and the second surface, and a spacer portion (50) provided to be located at least between the connection substrate and the first qubit substrate, wherein the first surface and the second surface are located in a same plane, and the spacer portion is provided to be in contact with a third surface (15a) that is provided in a peripheral portion of the first qubit substrate and is lower than the first surface.

A method of training a neural network as a natural language processing, NLP, model, comprises: inputting annotated training data to first architecture portions of the neural network, the first architecture portions being executed respectively in a plurality of distributed client computing devices in communication with a server computing device, the training data being derived from text data private to the client computing device in which the first architecture portion is executed, the server computing device having no access to any of the private text data; deriving from the training data, using the first architecture portions, weight matrices of numeric weights which are decoupled from the private text data; concatenating the weight matrices, in a second architecture portion of the neural network executed in the server computing device, to obtain a single concatenated weight matrix; and training, on the second architecture portion, the NLP model using the concatenated weight matrix.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

A quantum calculation support program for causing a computer to execute processing including: expanding an imaginary-time evolution expression into multiple expressions for multiple orders; generating sets of orders obtained by extracting an order twice from the multiple orders; performing, for each set, generating a quantum circuit indicating quantum calculation of the physical quantity obtained by partial imaginary-time evolution using expressions of orders in the set, and causing a quantum computer to repeatedly execute the quantum calculation using the quantum circuit for the set until the physical quantity obtained from the quantum calculation converges such that an output state obtained by the quantum calculation using the quantum circuit for the set is used as an input state in the subsequent quantum calculation; and calculating the thermal equilibrium expectation value at the finite temperature of the physical quantity using a value of the physical quantity after convergence for each set.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …