Abstract: A virtual reality (VR) sickness assessment method according to an embodiment includes receiving virtual reality content, and quantitatively evaluating virtual reality sickness for the received virtual reality content using a neural network based on a pre-trained neural mismatch model. The evaluating of the virtual reality sickness may include predicting an expected visual signal for an input visual signal of the virtual reality content based on the neural mismatch model, extracting a neural mismatch feature between the predicted expected visual signal based on the neural mismatch model and an input visual signal for a corresponding frame of the virtual reality content corresponding to the expected visual signal, and evaluating a level of the virtual reality sickness based on the neural mismatch model and the extracted neural mismatch feature.

Abstract: A method for searching deep neural network architecture for computation offloading in a computing environment is provided.

Abstract: Disclosed are an image processing method and device using a line-wise operation. The image processing device, according to one embodiment, comprises: a receiver for receiving an image; a first convolution operator for generating a feature map by performing a convolution operation on the basis of the image; and a compressor for compressing the feature map into units of at least one line; and a decompressor for reconstructing the feature map compressed into units of lines.

Abstract: Proposed are an optimal operation method of a high-frequency dithering technique for compensating for interference noise in an analog optical transmission-based mobile fronthaul network, and a transmitter using same. An interference noise compensation method using high-frequency phase dithering performed in an analog optical transmission-based mobile fronthaul network may include the steps in which: a frequency-multiplexed wireless signal is converted in an optical transmitter to an intensity-modulated optical signal; and the phase of the optical signal intensity-modulated in the optical transmitter is dithered with an Orthogonal Frequency-Division Multiplexing (OFDM) signal.

Abstract: Embodiments of the present disclosure are directed to a diesel reformer system comprising: a diesel autothermal reforming unit; a post-reforming unit disposed downstream of the autothermal reforming unit; a heat exchanger disposed downstream of the post-reforming unit; and a desulfurization unit disposed downstream of the heat exchanger.

Abstract: A meta input method and system and a user-centered inference method and system via a meta input for recycling of a pretrained deep learning model are provided. The meta input method for the recycling of the pretrained deep learning model performed by a computer device includes optimizing a meta input by considering a relation between input data and output prediction of the pretrained deep learning model and adding the optimized meta input to testing data in a user environment to transform distribution of the testing data into distribution of training data used to build the deep learning model.

Abstract: Disclosed herein are a floating-point computation apparatus and method using Computing-in-Memory (CIM).

Abstract: A method of manufacturing self-healing polymer capable of controlling physical properties is provided. The method includes forming the self-healing polymer by adjusting a copolymer composition using monomers of glycidyl methacrylate (GMA) and 2-hydroxyethyl acrylate (HEA) and an initiator of tert-butyl peroxide (TBPO) based on an initiated chemical vapor deposition method (iCVD).

Abstract: An interface for optical communication, including an input waveguide in which light input from an outside is guided, an output waveguide including a first part abutting against one end of the input waveguide and a second part connected to the first part and a substrate including a Buried oxide (BOX) layer connected to a lower side of the output waveguide, wherein the one end of the input waveguide includes a tapered structure of which a cross-sectional area is reduced by a predetermined angle.

Abstract: An information processing apparatus comprises a controller. The controller is configured to execute: acquiring a plurality of datasets, each of the datasets being configured with a combination of training data and a correct answer label; and implementing machine learning of an estimation model using the acquired plurality of datasets, wherein the training data includes workload information about an application constructed based on a microservice architecture and resource use information about resources used for each of components included in the application, in a learning target environment, the correct answer label is configured to show a true value of quality of service of the application, and the machine learning comprises training the estimation model such that, for each of the datasets, an estimated value of the quality of service calculated with the estimation model based on the training data corresponds to the true value shown by the correct answer label.

Abstract: A processor-implemented light source information output method includes: receiving an input image; detecting, using a trained neural network, at least one object in the input image; estimating, using the trained neural network, light source information of a light source corresponding to the at least one object; and outputting the light source information.

Abstract: A method and apparatus for modulating/demodulating an FSK signal capable of overcoming a trade-off relationship between a modulation index and a spectral efficiency are disclosed. An apparatus for modulating/demodulating a frequency deviation keying (FSK) signal includes a channel selection-modulator, a phase locked loop, and an output unit. The channel selection-modulator modulates an FSK signal by setting a frequency channel to be used. The phase locked loop generates a desired output frequency ‘fout’ compared to a reference frequency ‘fREF’ by adjusting a frequency division ratio (N+n) with respect to a frequency of the modulated FSK signal. The output unit amplifies the FSK signal having the generated output frequency ‘fout’ and radiating the amplified FSK signal through an antenna. Here, each of the frequency channels is divided into two or more tones, and different frequency channels are allocated between the tones divided into two or more tones.

Abstract: Provided are a graphene-based fiber in which a liquid-crystalline aromatic compound is intercalated into a graphene-based material, a graphene-based carbon fiber obtained by carbonizing the graphene-based fiber, and a method of manufacturing the same.

Abstract: Disclosed herein is a system for social control and use of an Internet of Things (IoT) device, comprising an actuator, one or more mobile devices, and a control server. The actuator is arranged or disposed in a public or common space. The one or more mobile devices comprises a social control user interface (UI), which includes a voting function. The control server comprises a preference aggregation engine for deriving a consensus by aggregating vote(d) values from the one or more mobile devices, and a device control command generating unit for generating a device control command based on the consensus. Other embodiments are described and shown.

Abstract: According to an embodiment of the present disclosure, a neuron circuit may be provided. The neuron circuit includes a biristor that includes a collector electrode receiving a constant input current from a first synapse circuit and an emitter electrode connected with a ground and outputs a collector signal through the collector electrode, and a voltage divider that is enabled by the collector signal, performs voltage division on an operating voltage by using values of resistances included therein, and outputs an output voltage corresponding to a result of the voltage division to a second synapse circuit.

Abstract: A method for generating a trained model is provided. The method for generating a trained model includes: receiving a learning data; generating an asymmetric multi-task feature network including a parameter matrix of the trained model which permits an asymmetric knowledge transfer between tasks and a feedback matrix for a feedback connection from the tasks to features; computing a parameter matrix of the asymmetric multi-task feature network using the input learning data to minimize a predetermined objective function; and generating an asymmetric multi-task feature trained model using the computed parameter matrix as the parameter of the generated asymmetric multi-task feature network.

Abstract: Spatio-cohesive service discovery and handover methods for distributed IoT environments are disclosed. The spatio-cohesive service discovery and handover methods include discovering a set of IoT (internet of thing) resources for providing a set of services which is a set of functionalities necessary for a task in distributed IoT environments, wherein the discovering a set of IoT resources may discover the set of IoT resources through a spatio-cohesive method considering spatial distance between a user and a service and between two services.

Abstract: Proposed is a federated learning system. The federated learning system comprises: a central server configured to transmit at least one global parameter of a global model to each client device, receive at least one local parameter of a local model trained from each of client devices, and update the global model using the at least one local parameter; and a plurality of client devices configured to train the local model by applying a loss between a predicted value of the global model and a predicted value of the local model possessed by itself to a loss function, and transmit at least one local parameter of the trained local model to the central server.

Abstract: Proposed is a federated learning system. The federated learning system may comprise: a central server configured to transmit a first parameter of an extractor in a federated learning model including the extractor and a classifier to each of a plurality of client devices, and receive a plurality of first parameters learned from the plurality of client devices to update the federated learning model; and the plurality of client devices configured to train each of the plurality of the first parameters of the federated learning model using a training data set stored in each of the plurality of client devices while maintaining a value of a second parameter value of the classifier in the federated learning model, and to transmit each of the plurality of the trained first parameters to the central server.

Abstract: An apparatus and a method for the disclosure relates to an artificial intelligence (AI) system that simulates functions such as recognition and determination of the human brain by using a machine training algorithm such as deep learning and an application of the AI system are provided. A neural network training method includes obtaining target modality signals of a first domain aligned in a time order and auxiliary modality signals of a second domain that are not aligned in the time order, extracting characteristic information of the target modality signals using a first neural network model, estimating the time order of the auxiliary modality signals using a second neural network model, and training the first neural network model based on a result of the estimation and the characteristic information.