Abstract: An apparatus for accelerating neural networks, includes: a memory for storing graph input data including vertices and edges; an aggregation engine that processes the accumulation of features and generates feature vectors by taking the graph input data and performing an aggregation operation on the graph input data; an on-chip cache for caching the feature vectors; and a combination engine that generates a systolic array for matrix multiplications based on the feature vectors taken from the on-chip cache and weights taken from the memory.

Abstract: A broadband patch antenna comprises a substrate, a ground plate attached to one surface of the substrate, a radiation plate attached to the other surface of the substrate opposite the one surface of the substrate, and a feed line attached to the other surface of the substrate and having one end connected to the radiation plate. The feed line comprises a first line and a second line. The ground plate has the shape of capital “L” having a first groove, a second groove, and a third groove, and may not comprise a portion corresponding to the radiation plate. The first groove is positioned at a first portion which corresponds to a connection portion between the first line and the radiation plate, the second groove is positioned at a second portion which corresponds to a connection portion between the second line and the radiation plate, and the third groove may be positioned so as to be spaced apart from the first groove and the second groove.

Abstract: A method for an anomaly detection is provided. The method may include acquiring a score predictor trained using normal time-series data, wherein the score predictor is a deep learning model configured to output a conditional score for previous time-series data and the conditional score represents a gradient of data density, extracting data for a specific time and data segments corresponding to a period before the specific time from target time-series data, and predicting a conditional score for the data segments through the trained score predictor and conducting an anomaly determination for the data for the specific time using the predicted conditional score.

Abstract: Provided are: a peptide targeting Fusobacteria, comprising a peptide consisting of an amino acid sequence selected from the group consisting of SEQ ID NOs: 77 to 83; a composition for diagnosing cancer, comprising same; and a drug delivery composition. The peptide targeting Fusobacteria accurately targets Fusobacteria, which are harmful bacteria in various diseases, including cancer, and thus can be effectively used in cancer diagnosis and drug delivery.

Abstract: An Enhanced Programmable Addition via Site-specific Targeting Elements (EPASTE) system utilizing a fusion protein of Prime Editor (PE) protein and Pa01 Integrase, and the EPASTE system enables the integration of large donor DNA into a target site of an intracellular genome with significantly superior efficiency than existing technologies, and without the need for a DNA repair process for DNA double-strand breakage.

Abstract: A memory system includes a memory resource and a smart controller. The memory resource includes semiconductor memory devices, the semiconductor memory devices are divided into a first semiconductor memory and a second semiconductor memory for each of a plurality of channels, and the first semiconductor memory the second semiconductor memory belong to different ranks. The smart controller, connected to the semiconductor memory devices through the channels, controls the semiconductor memory devices by communicating a plurality of hosts through a compute express link (CXL) interface, and each of the plurality of hosts drives at least one virtual machine. The smart controller controls a power mode of the memory resource by managing an idle memory region from among a plurality of memory regions of the plurality of semiconductor memory devices at a rank level without intervention of the plurality of hosts, the plurality of memory regions storing data.

Abstract: An embedded memory device includes a plurality of first bit cells configured to store data and connected between a first bitline and a first complementary bitline, and at least one first cropping cell connected between the first bitline and the first complementary bitline. The at least one first cropping cell electrically connects a global bitline to the first bitline and electrically connects a complementary global bitline to the first complementary bitline in response to a first crop wordline signal. The global bitline and the complementary global bitline are implemented as an upper metal member, and the first bitline and the first complementary bitline are implemented as a lower metal member disposed below the upper metal member.

Abstract: The present disclosure relates to a silicon/carbon anode composite for a lithium secondary battery, a method for preparing the same and a lithium secondary battery including the same. Particularly, the silicon/carbon anode composite for a lithium secondary battery is obtained by coating the surface of silicon microparticles with a polymer or pitch and allowing the resultant product to be physicochemically bound to the surface of the carbonaceous material through a mechanofusion process, and thus needs no additional heat treatment and can be processed with ease, is prevented from volumetric swelling of silicon caused by lithium-ion intercalation and deintercalation, and can provide a battery with significantly improved capacity and life characteristics.

Abstract: An apparatus for inspecting an antibiotic and a method for determining antibiotic sensitivity using the same is provided. The antibiotic susceptibility inspection time which has conventionally taken longer than 24 hours is shortened to about 2 hours or less, the efficacy of the target substance is monitored in real time, the identification of the microorganism, the kind of the antibiotic capable of treating the microorganism, and the minimum dosage thereof are quickly confirmed. Microbial infections requiring prompt diagnosis and treatment can be effectively treated.

Abstract: Disclosed herein is a method of controlling power of a base station in a spectrum sharing environment, which includes setting, by a control device, predetermined initial power information for a plurality of base stations located in a service area in a time slot t; calculating, by the control device, a coverage radius for each of the base stations on the basis of the initial power information; calculating, by the control device, gradients of a coverage area and an interference area for each of the base stations on the basis of the coverage radius for each of the base stations; and updating, by the control device, power information for each of the base stations in a time slot t+1 on the basis of the gradients of each of the base stations.

Abstract: The present disclosure is directed to providing a method for preparing powdered cultured meat, which can increase the efficiency of cell differentiation and the protein content of cultured meat with economic feasibility and to providing powdered cultured meat with high physicochemical stability and transportability and a food composition containing the same. The powdered cultured meat has a flavor similar to that of real meat, and can be highly utilized as a high-protein food in a variety of food compositions or processed foods.

Abstract: Proposed is a dual-light-emitting material that includes a porous framework composed of a metal ion and an organic ligand, an insert body placed in a cavity of the porous framework, and a nanocrystal containing the metal ion of the porous framework, wherein the organic ligand is configured to emit room-temperature organic phosphorescence (RT-OP) and the nanocrystal is configured to emit fluorescence.

Abstract: A method and apparatus for encoding/decoding audio signal are provided. The encoding method includes transforming an input audio signal in a time domain into an audio signal in a frequency domain, quantizing energy of a frequency band of the audio signal in the frequency domain, generating a normal signal by normalizing the audio signal in the frequency domain according to quantized energy, obtaining a feature vector including information on the energy of the frequency band based on the normal signal and the input audio signal, quantizing the feature vector, obtaining a scale factor used to scale the normal signal based on the quantized feature vector, quantizing an adjustment signal into which the normal signal has been scaled based on the scale factor, and outputting bitstreams based on the quantized energy, the quantized feature vector, and the quantized adjustment signal.

Abstract: An artificial intelligence (AI) device based on a trust environment, includes a first type memory configured to transmit encrypted input data and receive encrypted output data, and a trust AI processing unit configured to operate in a trust space and perform AI computation of the encrypted input and output data. The trust AI processing unit includes: a cryptographic processing front-end processor configured to generate decrypted input data through decryption of the encrypted input data and perform encryption of non-encrypted output data to generate the encrypted output data, a second type memory configured to provide a buffer for the decrypted input data and the non-encrypted input data, and a processor configured to perform a neural network computation based on the decrypted input data to generate the non-encrypted output data.

Abstract: A battery pack housing for improved cooling efficiency of a battery cell is disclosed. A cooling fluid introduced into a battery pack of the battery pack housing flows to generate turbulence around a battery tab at which high-temperature heat is typically generated. A flow rate and a flow velocity of the cooling fluid may be increased around the battery tab, the cooling fluid may be more actively circulated, and a temperature deviation between cells in the battery pack may be effectively reduced so that performance degradation of the battery pack may be prevented and a lifetime of the battery pack may be increased. In addition, even when the same amount of cooling fluid is used, the cooling efficiency of the battery pack may be improved, and an amount of the fluid needed for cooling may be reduced to simplify an entire system of the battery pack.

Abstract: The present disclosure relates to a thick film electrode and a method for manufacturing the same. Since the thick film electrode uses a photocurable monomer as a processing solvent, it requires no separate drying step and may have a sufficiently large thickness without physical cracking. In addition, since the thick film electrode uses a nonwoven web-type substrate having a carbon body applied to the surface of a conductive polymer, it may have higher ion conductivity and areal capacity. Further, a semi solid-state battery including the thick film electrode may have excellent energy density.

Abstract: A method for providing a feedback on a surgical outcome by a computer includes dividing, by the computer, actual surgical data obtained in an actual surgical process into a plurality of detailed surgical operations to obtain actual surgical cue sheet data composed of the plurality of detailed surgical operations, obtaining, by the computer, reference cue sheet data about the actual surgery, and comparing, by the computer, the actual surgical cue sheet data with the reference cue sheet data, and providing, by the computer, the feedback based on the comparison result.

Abstract: A processing-in-memory (PIM)-based computing device includes a main memory configured to store first and second data sets, a PIM module configured as an in-line memory module including a memory processor, a working memory, and a memory bank, the memory processor being connected to the working memory and performing a join operation on the first and second data sets loaded into the working memory, and a processor configured to form first and second channels with the main memory and the PIM module, import the first and second data sets through the first channel, and store the first and second data sets in the memory bank through the second channel.

Abstract: The present invention relates to an autoimmune disease animal model and a method for preparing the same. The animal model of the present invention efficiently generates an autoantibody that causes autoimmune disease, by introducing an autoantigen-encoding gene through the Cre-LoxP system, which enables stable and continuous expression of the autoantigen at a desired time by treatment of Cre recombinase. Contrary to the conventional method that fails to effectively produce autoantibody due to leaky expression and immune tolerance, or requires repeated administration over several weeks, the present invention achieves sustainable expression of an effective amount of autoantigen by only one TAT-Cre recombinase treatment, thus may be utilized as an outstanding animal model reproducing various symptoms and molecular mechanisms of autoimmune diseases.

Abstract: Provided is a control method of a deep neural network (DNN) accelerator for optimized data processing. The control method includes, based on a dataflow and a hardware mapping value of neural network data allocated to a first-level memory, calculating a plurality of offsets representing start components of a plurality of data tiles of the neural network data, based on receiving an update request for the neural network data from a second-level memory, identifying a data type of an update data tile corresponding to the received update request among the plurality of data tiles, identifying one or more components of the update data tile, based on the data type of the update data tile and an offset of the update data tile among the calculated plurality of offsets, and updating neural network data of the identified one or more components between the first-level memory and the second-level memory.