Abstract: A porous transport layer is disclosed. The porous transport layer includes a base layer containing a titanium family element, a first coating layer disposed on one surface of the base layer and containing iridium (Ir), and a second coating layer disposed on the other surface of the base layer, and containing at least one of platinum (Pt), gold (Au), and silver (Ag), and a method for preparing the same.

Abstract: Provided are a method for generating a summary and a system therefor. The method according to some embodiments may include calculating a likelihood loss for a summary model using a first text sample and a first summary sentence corresponding to the first text sample, calculating an unlikelihood loss for the summary model using a second text sample and the first summary sentence, the second text sample being a negative sample generated from the first text sample, and updating the summary model based on the likelihood loss and the unlikelihood loss.

Abstract: Provided is an aerogel blanket and a method for producing the same, wherein a catalyzed sol is sufficiently and uniformly impregnated into a blanket in an impregnation tank, and the catalyzed sol is allowed to stay in the impregnation tank for a specific time to control fluidity while achieving a viscosity at which the catalyzed sol can be easily introduced into the blanket, thereby forming a uniform aerogel in the blanket. As a result, the uniformity of pore structure and thermal insulation performance of an aerogel blanket are improved, the loss of raw materials is reduced through the impregnation process, the occurrence of process problems is reduced, and the generation of dust is reduced.

Abstract: An apparatus for embedding a sentence feature vector according to an embodiment includes a sentence acquisitor configured to acquire a first sentence and a second sentence, each including one or more words; a vector extractor configured to extract a first feature vector corresponding to the first sentence and a second feature vector corresponding to the second sentence by independently inputting each of the first sentence and the second sentence into a feature extraction network; and a vector compressor configured to compress the first feature vector and the second feature vector into a first compressed vector and a second compressed vector, respectively, by independently inputting each of the first feature vector and the second feature vector into a convolutional neural network (CNN)-based vector compression network.

Abstract: Provided is an aerogel blanket and a method for producing the same, wherein a catalyzed sol I sufficiently and uniformly impregnated into a blanket in an impregnation tank, and the catalyzed sol is allowed to stay in the impregnation tank for a specific time to control fluidity while achieving a viscosity at which the catalyzed sol can be easily introduced into the blanket, thereby forming a uniform aerogel in the blanket. As a result, the uniformity of pore structure and thermal insulation performance of an aerogel blanket are improved, the loss of raw materials is reduced through the impregnation process, the occurrence of process problems is reduced, and the generation of dust is reduced.

Abstract: According to an embodiment, a cordless ultrasonic surgery instrument comprises an end effector directly contacting a tissue for incision or suture of the tissue, a handpiece including at least one or more piezoelectric elements to transfer vibration energy to the end effector, a hand instrument including a controller controlling an operation of the end effector and controlling to increase power from a power source device, a shaft assembly coupled with the end effector, rotated by receiving the vibration energy from the handpiece, converting part of vibration energy according to the vibration into electrical energy, and collecting the electrical energy, and a battery supplying power to the handpiece to allow the handpiece to transfer the vibration energy to the end effector, the battery charged with the electrical energy transmitted from the shaft assembly.

Abstract: A method for a tournament-type on-line game by batch processing includes generating an independent game turn determining both participants in each match of the game, to every participant; providing match information, which includes the game turn information, victory criterion information determined in association with the game turn information and allowing the participant to win in each match until stage K, and an input circumstance for inputting variable information, to a client of the participant; receiving variable information from the client through the provided match information; and operating all of the received variable information by batch processing to determine winners until stage K, wherein an operation result of the variable information received from both clients of each match is compared with the victory criterion information so that the participant having the victory criterion information identical to the operation result is determined as a winner of each match.

Abstract: One aspect of the invention provides a method for training a classification algorithm to detect a retinal pathology. The method includes: for a plurality of pseudo two-dimensional data sets of one-dimensional data points, each pseudo two-dimensional data point representing RNFL thickness values for a subject and corresponding index values for the data points: performing fractal analysis on the data set to calculate a plurality of fractal dimensions and calculating a plurality of slopes between each fractal dimension; combining the plurality of slopes for subjects labeled as pathologic into a pathologic data set; combining the plurality of slopes for subjects labeled as healthy into a healthy data set; and applying a linear discriminant function the pathologic data set and the healthy data set; thereby training a classification algorithm to detect the retinal pathology.

Abstract: A method of estimating a read level for a memory device includes calculating first information corresponding to at least one among information about the number of cells having a particular logic level in data to be programmed and information about the number of cells having a particular cell state and storing the first information during a program operation; reading the data based on a threshold level that has been set and calculating second information about the number of cells in at least one state defined by the threshold level with respect to the read data; calculating third information about the number of cells in the at least one state, which corresponds to the second information, using a probability based on the first information; comparing the second information with the third information; and determining whether to change the threshold level according to the comparison result.

Abstract: A method for a tournament-type on-line game by batch processing includes generating an independent game turn determining both participants in each match of the game, to every participant; providing match information, which includes the game turn information, victory criterion information determined in association with the game turn information and allowing the participant to win in each match until stage K, and an input circumstance for inputting variable information, to a client of the participant; receiving variable information from the client through the provided match information; and operating all of the received variable information by batch processing to determine winners until stage K, wherein an operation result of the variable information received from both clients of each match is compared with the victory criterion information so that the participant having the victory criterion information identical to the operation result is determined as a winner of each match.

Abstract: A method of estimating a read level for a memory device includes calculating first information corresponding to at least one among information about the number of cells having a particular logic level in data to be programmed and information about the number of cells having a particular cell state and storing the first information during a program operation; reading the data based on a threshold level that has been set and calculating second information about the number of cells in at least one state defined by the threshold level with respect to the read data; calculating third information about the number of cells in the at least one state, which corresponds to the second information, using a probability based on the first information; comparing the second information with the third information; and determining whether to change the threshold level according to the comparison result.