Abstract: Disclosed are an apparatus for implementing hardware-based dropout for an artificial neural network using a selector element and a neural network circuit system using the same, and an apparatus for implementing hardware-based dropout for an artificial neural network using a selector element according to an exemplary embodiment of the present disclosure may include: a switch unit disposed on the bit line of a hardware-based artificial neural network, and provided to allow a target current which flows to the bit line to selectively flow to a predetermined neuron or ground node constituting the artificial neural network; and a driving unit which includes a selector element turned on or off according to a size relationship of an applied driving voltage and a threshold voltage, and controlling the switch unit.

Abstract: A coacervate including a catechol-substituted anionic polymer; an adhesive including same; and a method for producing the coacervate are described. More specifically, a coacervate formed by mixing a catechol derivative of a mussel adhesive protein and a catechol-substituted anionic polymer; an adhesive including the coacervate; and a method for producing a coacervate are described. The method includes a step of obtaining a catechol-substituted anionic polymer through catechol substitution of an anionic polymer, and a step of mixing the catechol-substituted anionic polymer and a catechol derivative of a mussel adhesive protein.

Abstract: A deep learning-based cooling system temperature prediction apparatus has an artificial neural network modeled by connecting a plurality of artificial neural network submodels each including an input layer, a hidden layer, and an output layer is used. A pump flow speed, a cooling water flow rate, a battery inlet cooling water temperature, a motor inlet cooling water temperature, a radiator outlet cooling water temperature, a battery temperature, and a motor temperature are predicted by inputting at least one of a predetermined control variable, an environment variable, or a time variable to the plurality of artificial neural network submodels in accordance with a physical causality. A number of the plurality of artificial neural network submodels and the control variables or environment variables that are sequentially input to each submodel depend on divisional control and integral control of the cooling system.

Abstract: Proposed is a vibration sensor including: a substrate; a first electrode positioned on the substrate; a support positioned on the first electrode and including a cylindrical hollow hole; and a diaphragm including a thin film positioned on the support and a second electrode positioned on the thin film. According to the present disclosure, it is possible to manufacture a skin-attachable vibration sensor that is attached to a user's neck to detect vibration acceleration in user's neck skin, thus exhibiting a uniform and high sensitivity to a user's voice over the frequency range of the human voice. In addition, the sensor sensitively detects a user's voice through neck skin vibrations rather than through air, thus being free from the influence of external noise or wind, and can recognize the user's voice even in a situation where a user's mouth is covered.

Abstract: Provided is a neural processing device including a noise classifier configured to perform a preprocessing on input data to determine a noise characteristic of the input data, a network selector configured to select one of a plurality of neural networks based on the noise characteristic, and an operator configured to perform inference on the input data based on selected weights corresponding to the selected neural network.

Abstract: A photoacoustic, noninvasive, and continuous blood glucose measurement device includes: a diode laser for irradiating a living body with a pulsed laser signal having a specific wavelength; and an ultrasound transducer for measuring a photoacoustic signal in a form of ultrasonic waves generated by a reaction of the living body with the pulsed laser signal.

Abstract: A user authentication method performed by an authentication management server constituting a block chain network with a target device may comprise receiving, from the target device, a transaction including authentication information of a user and a unique value of the target device; determining a token of the user stored in a smart contract based on the authentication information; and authenticating the user based on validity information of the token, ownership information of the token, and the transaction.

Abstract: Porous carbon particles, and a positive electrode active material and a lithium secondary battery including the same. This may improve the energy density of the lithium secondary battery by applying a porous electrode containing micropores and mesopores and having a uniform size distribution and shape as a positive electrode material.

Abstract: Disclosed herein is a transparent ultrasonic sensor-based ultrasonic optical composite imaging system including at least one light source and a transparent ultrasonic sensor coaxially aligned with light emitted from the light source and allowing the light emitted from the light source to be transmitted therethrough, wherein the transparent ultrasonic sensor is disposed between the light source and a subject.

Abstract: The present disclosure relates to a GIS geographical information retrieval method using a dynamic k-nearest neighbor search algorithm in an obstacle environment, which has been devised to search for geographical information by using a dynamic k-nearest neighbor search algorithm in an obstacle environment. The GIS geographical information retrieval method using a dynamic k-nearest neighbor search algorithm in an obstacle environment has an effect in that it can optimize a search time by considering obstacles as a range of angles, not discrete objects, on the basis of a query point, using the shortest path characteristics at the same time, and assigning priority to neighbors based on the calculated range of angles.

Abstract: A method of implementing a neural network in a neuromorphic apparatus having a memory and processing circuitry, where the method includes dividing, by the processing circuitry, the neural network into a plurality of sub-networks based on a size of a core of the memory to implement the neural network, initializing, by the processing circuitry, a hyper-parameter used in the sub-networks, and training, by the processing circuitry, the sub-networks by using the hyper-parameter.

Abstract: A weight confirmation method for analog synaptic devices of an artificial neural network includes: learning artificial neural network hardware based on artificial analog memory devices using an artificial neural network learning algorithm; after the learning of the artificial neural network hardware, reading a total weight of a pair of synaptic devices; comparing the total weight of the pair of synaptic devices with 0; applying weights to a positive synaptic device and a negative synaptic device of the pair of synaptic devices, respectively; and confirming the total weight of the pair of synaptic devices in accordance with the weight of the positive synaptic device and the weight of the negative synaptic device. A retention problem of analog synaptic devices is overcome and thus the artificial neural network may stably operate.

Abstract: The present invention relates to a Bifidobacterium bifidum inducing regulatory T cells (Treg), a polysaccharide derived from Bifidobacterium bifidum, and a probiotic strain producing a polysaccharide and, more particularly, a polysaccharide containing ?-1-6-glucan as an effective ingredient, a probiotic strain producing ?-1-6-glucan, a food comprising the polysaccharide or strain as an effective ingredient for alleviation of immune disease or inflammatory disease, a therapeutic agent comprising the polysaccharide or strain as an effective ingredient for alleviation of immune disease or inflammatory disease, a method for preparing induced regulatory T cells (iTreg) by treatment with the polysaccharide or strain, and a cell therapy product for prevention or treatment of immune disease or inflammatory disease, comprising the induced regulatory T cells prepared by the method.

Abstract: A method formats a weight matrix in a current layer included in a neural network. The method includes calculating a row length for each row of the weight matrix based on a number of elements each of which has non-zero value; storing rearrangement information including result of sorting rows in the order of row lengths; performing a row transformation or the row transformation and a column transformation on the weight matrix using the rearrangement information; distributing rows of a transformed weight matrix to a plurality of processing elements (PEs); and generating formatted data including one or more group data each including values and column information being processed in each of the PEs.

Abstract: The present invention relates to a method for controlling an ultrafast chemical reaction using a microfluidic reactor, and more specifically, the present invention relates to a method for controlling an ultrafast chemical reaction such as the Fries rearrangement reaction and the like by using a microfluidic reactor by the 3D metal printing technique.

Abstract: The present disclosure provides technology that proposes an ultra-high frequency band (mmWave band) vertical polarization antenna having a new structure applicable to a slim planar structure (e.g., a terminal).

Abstract: Porous carbon particles, and a positive electrode active material and a lithium secondary battery including the same. This may improve the energy density of the lithium secondary battery by applying a porous electrode containing micropores and mesopores and having a uniform size distribution and shape as a positive electrode material.

Abstract: The present invention relates to a method for mass-producing a target protein in a plant and, more specifically, to a method for producing high levels of protein in plant leaf cells, in which an expression cassette construct for high expression of a target gene and an expression cassette construct for high expression of p38 as a gene silencing suppressor, are prepared, a binary vector for transient expression in plant cells, having the construct together, is then prepared, and the binary vector is then subjected to agrobacterium-mediated transformation by using a single agrobacterium culture.

Abstract: An upconversion nanoparticle includes at least one host selected from LiYF4, NaY, NaYF4, NaGdF4, and CaF3, at least one sensitizer selected from Sm3+, Nd3+, Dy3+, Ho3+, and Yb3+ doped in the at least one host, and at least one activator selected from Er3+, Ho3+, Tm3+, and Eu3+ doped in the at least one host. The upconversion nanoparticle is designed using a calculation from first principles to absorb light in the near-infrared wavelength range whose stability is ensured. Further, a hyaluronic acid-upconversion nanoparticle conjugate, in which the upconversion nanoparticle as described above is bonded to hyaluronic acid, is provided to be used in various internal sites with a hyaluronic acid receptor, particularly enables targeting, and increases an internal retention period and biocompatibility thereof.

Abstract: A method and an apparatus for performing a Superposition Coded Modulation (SCM) scheme in a broadcasting or communication system including a controller are provided. The method includes controlling an SCM coefficient by the controller. The SCM coefficient is controlled according to a channel capacity of each layer of one or more layers in which information included in a signal is encoded.