Abstract: An accelerator includes: a memory configured to store input data; a plurality of shift buffers each configured to shift input data received sequentially from the memory in each cycle, and in response to input data being stored in each of internal elements of the shift buffer, output the stored input data to a processing element (PE) array; a plurality of backup buffers each configured to store input data received sequentially from the memory and transfer the stored input data to one of the shift buffers; and the PE array configured to perform an operation on input data received from one or more of the shift buffers and on a corresponding kernel.

Abstract: A method and apparatus for classification using a neural network. A classification apparatus includes at least one processor and a memory. The memory is configured to store a classifier and a preprocessor including a defensive noise generator. The at least one processor generates defensive noise from an input image through the defensive noise generator in the preprocessor, generates a combined image by combining the input image and the defensive noise, and outputs a classification result by inputting the combined image to the classifier.

Abstract: A method of manufacturing a magnetic wire example includes depositing a magnetic film, which has a composition that enables measuring motion of a magnetic domain wall in the magnetic film, on/above a silicon substrate, forming the magnetic film on the silicon substrate on which the magnetic film is deposited using a wire pattern and an electrode pattern of a certain specification, shielding a central part of the magnetic wire in a photolithography method by an edge milling pattern which corresponds to a predetermined specification, and ablating an edge portion of the magnetic wire which is not shielded by an ion milling.

Abstract: An accelerator, an operation method of the accelerator, and an accelerator system including the accelerator are disclosed. The operation method includes receiving one or more workloads assigned on an accelerator, determining reuse data of the workloads based on hardware resource information and/or a memory access cost of the accelerator when a plurality of processing units included in the accelerator performs the workloads, and providing a result of performing the workloads.

Abstract: An authentication method and apparatus using a transformation model are disclosed. The authentication method includes generating, at a first apparatus, a first enrolled feature based on a first feature extractor, obtaining a second enrolled feature to which the first enrolled feature is transformed, determining an input feature by extracting a feature from input data with a second feature extractor different from the first feature extractor, and performing an authentication based on the second enrolled feature and the input feature.

Abstract: An accelerator, an operation method of the accelerator, and an accelerator system including the accelerator are disclosed. The operation method includes receiving one or more workloads assigned by a host controller, determining reuse data of the workloads based on hardware resource information and/or a memory access cost of the accelerator when a plurality of processing units included in the accelerator performs the workloads, and providing a result of performing the workloads.

Abstract: An authentication method and apparatus using a transformation model are disclosed. The authentication method includes generating, at a first apparatus, a first enrolled feature based on a first feature extractor, obtaining a second enrolled feature to which the first enrolled feature is transformed, determining an input feature by extracting a feature from input data with a second feature extractor different from the first feature extractor, and performing an authentication based on the second enrolled feature and the input feature.

Abstract: An encryption method and apparatus based on homomorphic encryption using an odd function property. The encryption method includes generating a ciphertext by encrypting data, and bootstrapping the ciphertext by performing a modular reduction based on an odd function property for a modulus corresponding to the ciphertext.

Abstract: A probe includes a working channel, at least one optical element provided in the working channel and configured to acquire a spectral image of surface information of an object, and an ultrasonic transducer arranged in the working channel and configured to acquire an ultrasonic image of depth information of the target and move with respect to the at least one optical element.

Abstract: Provided is a smart insole including a support layer; a plurality of pressure sensors provided to the support layer and configured to sense a pressure; a plurality of vibrators provided to the support layer and configured to generate a vibration; and a controller configured to determine a center of pressure (COP) based on a pressure sensed by each of the plurality of pressure sensors and to control the plurality of vibrators based on a positional relationship between a setting point and the COP.

Abstract: A device and method with transformer model implementation are provided. The electronic device includes a processor configured to perform an inference by implementing a transformer model including a plurality of encoders and a plurality of decoders, and a memory configured to store instructions to be executed by the processor. Each of the encoders and the decoders includes an attention block that determines an attention value. The processor is configured to perform a first sub-softmax tile-wise operation in the attention block, perform a reduction operation to determine an adjustment factor based on a resulting value of the first sub-softmax operation, and perform a second sub-softmax tile-wise operation based on a resulting value of the reduction operation.

Abstract: A processor-implemented encryption method using homomorphic encryption includes: receiving data; generating a ciphertext by encrypting the received data; determining a coefficient of an approximating polynomial for performing a modular reduction on a modulus corresponding to the ciphertext, based on an error between the approximating polynomial and a modular reduction function; and performing bootstrapping on the ciphertext by performing the modular reduction based on the determined coefficient of the approximating polynomial.

Abstract: A neuromorphic memory circuit includes a plurality of memory cells, and each of the plurality of memory cells includes a first switching element having a threshold switching time determined based on a voltage applied to both ends of the first switching element at a time of receiving an input signal, and outputting the input signal in response to an elapse of the threshold switching time from a point in time at which the input signal is received; a first resistive memory element connected to the first switching element to divide the voltage applied to both ends of the first switching element; and a synapse circuit to generate an output signal in response to the input signal delayed by the threshold switching time.

Abstract: A method includes generating, by a neural network having a plurality of layers, final feature vectors of one or more frames of a plurality of frames of an input video, while sequentially processing each of the plurality of, and generating image information corresponding to the input video based on the generated final feature vectors. For each of the plurality of frames, the generating of the final feature vectors comprises determining whether to proceed with or stop a corresponding sequenced operation through layers of the neural network for generating a final feature vector of a corresponding frame, and generating the final feature vector of the corresponding frame in response to the corresponding sequenced operation completing a final stage of the corresponding sequenced operation.

Abstract: The present invention relates to a S1PR4-targeting composition for preventing or treating non-alcoholic steatohepatitis and, more particularly, to a pharmaceutical composition and a health functional food composition, both comprising a sphingolipid compound which serves as a functional inhibitor against S1PR4, showing prophylaxis and therapy of non-alcoholic steatohepatitis. The sphingolipid compound of the present invention is expected to be applied as a leading material effective for the prevention or treatment of non-alcoholic steatohepatitis (NASH) as it has the effect of reducing the infiltration of inflammatory cells into hepatic tissues and suppressing fibrosis and decreases a level of liver injury (ALT), inflammation in hepatic tissues, the expression of a fibrosis-related gene.

Abstract: The present disclosure relates to a 2-dimensional polymer nanosheet, a device including the nanosheet and a method of morphologically tunable preparing the nanosheet. Two-dimensional (2D) polymer nanosheets have been attracting immense attention owing to their potential applications in optical devices, membranes, and catalysis. A new crystalline polyacetylene is described that contains fluorenes and triisopropylsilyl side chains, which could self-assemble into sharp-edged 5-nm-thick square nanosheets with a narrow length dispersity of 1.01, by simple heating and aging in dichloromethane (DCM). The addition of tetrahydrofuran (THF) or chloroform to the heated polymer solution in DCM changed the morphology from square to rectangle. The aspect ratios increased linearly, from 1.0 to 10.6, according to the amount of THF or chloroform added, while maintaining narrow length dispersities less than 1.06.

Abstract: An encryption method and apparatus based on homomorphic encryption using a composition of functions. The encryption method includes generating a ciphertext by encrypting data, and bootstrapping the ciphertext by performing a modular reduction based on a composition of a function for a modulus corresponding to the ciphertext.

Abstract: A processor-implemented method of detecting a moving object includes: estimating a depth image of a current frame; determining an occlusion image of the current frame by calculating a depth difference value between the estimated depth image of the current frame and an estimated depth image of a previous frame; determining an occlusion accumulation image of the current frame by adding a depth difference value of the occlusion image of the current frame to a depth difference accumulation value of an occlusion accumulation image of the previous frame; and outputting an area of a moving object based on the occlusion accumulation image.

Abstract: An accelerator includes: a memory configured to store input data; a plurality of shift buffers each configured to shift input data received sequentially from the memory in each cycle, and in response to input data being stored in each of internal elements of the shift buffer, output the stored input data to a processing element (PE) array; a plurality of backup buffers each configured to store input data received sequentially from the memory and transfer the stored input data to one of the shift buffers; and the PE array configured to perform an operation on input data received from one or more of the shift buffers and on a corresponding kernel.

Abstract: The present invention provides a water-soluble fluorescent compound of resveratrone 6-O-?-glucoside [(E)-4-(8-hydroxy-6-(((2S,3R,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)oxy)naphthalen-2-yl)but-3-en-2-one] and its derivatives of Formula I which have at least one water-soluble substituent, and a method for preparing the same by a photochemical reaction of resveratrol 3-O-?-glucoside and its derivatives of having Formula 3 which are not fluorescent. Said new water-soluble fluorescent compounds has single-photon absorptive characteristics and/or two-photon absorptive characteristics as well as no or little toxicity, and can be usefully utilized in fields that requires water-soluble fluorescent characteristics (diagnosis, fluorescent probe, in vivo imaging, display, etc.).