Abstract: A learning device according to the present invention includes, in semi-supervised learning using domain information: a memory; and a processor. The processor performs operations. The operations includes: including a first neural network outputting data after predetermined conversion by using first data including the domain information and second data not including the domain information, a second neural network outputting a result of predetermined processing by using data after the conversion, and a third neural network outputting a result of domain discrimination by using data after the conversion; calculating a first loss being a loss of the domain discrimination; calculating a second loss being an unsupervised loss; calculating a third loss in the predetermined processing; and modifying a parameter of each of the first neural network to the third neural network in such a way as to decrease the second loss and the third loss and increase the first loss.

Abstract: A substrate processing system is disclosed which includes a substrate input/output chamber coupled to a transfer chamber, and one or more processing chambers coupled to the transfer chamber, wherein the substrate input/output chamber includes a plurality of stacked carrier holders, and a platen, wherein the platen includes a plurality of alignment pads and a plurality of openings formed in a recessed flange along a peripheral edge of the platen.

Abstract: An air intake apparatus for an internal combustion engine, the air intake apparatus includes: an air intake apparatus main body portion introducing air into a cylinder; a port portion provided integrally with a downstream side end portion of the air intake apparatus main body portion and inserted into an air intake port in a cylinder head; an air intake passage formed inside the air intake apparatus main body portion and the port portion, and through which an air-fuel mixture containing air and fuel flows; an injector provided in the air intake apparatus main body portion and introducing the fuel into the air intake passage; and a heater heating the fuel introduced by the injector. The injector is disposed at a position where the fuel is capable of being injected into the heater.

Abstract: A large amount of training data is typically required to perform deep network leaning, making it difficult to achieve using a few pieces of data. In order to solve this problem, the neural network device according to the present invention is provided with: a feature extraction unit which extracts features from training data using a learning neural network; an adversarial feature generation unit which generates an adversarial feature from the extracted features using the learning neural network; a pattern recognition unit which calculates a neural network recognition result using the training data and the adversarial feature; and a network learning unit which performs neural network learning so that the recognition result approaches a desired output.

Abstract: A composite material component manufacturing method includes a molding step of applying, by a pressing device, heat and pressure to a prepreg to mold the prepreg, and a transfer step of transferring the prepreg. The transfer step includes moving the prepreg together with the pressing device in a state in which a pressing force is applied to the prepreg by the pressing device.

Abstract: Embodiments described herein generally provide improved silicon-on-insulator (SOI) substrates for RF integrated circuits and methods for fabricating the same. In one embodiment, a method for forming a multilayer structure is provided. The method includes forming an intrinsic silicon layer on a handle substrate in a process chamber, wherein the handle substrate has a first resistivity and the intrinsic silicon layer having a second resistivity higher than the first resistivity, forming a charge trapping layer on the intrinsic silicon layer in the process chamber, bonding the handle substrate to an active substrate comprising a device layer and an insulating layer, wherein the insulating layer of the active substrate is in physical contact with the charge trapping layer, and removing at least a portion of the handle substrate.

Abstract: One implementation provides a method including providing a substrate into a processing chamber through a loading port, rotating the substrate clockwise, providing a gas mixture into a processing region through an inject insert comprising a first, second, and third sets of inject inlets, wherein the first set of inject inlets creates an inner zone inside the processing region, the second set of inject inlets creates a middle zone radially outward of the inner zone, and the third set of inject inlets creates an outer zone radially outward the middle zone, the gas mixture is provided by flowing the gas mixture through the first and second sets of inject inlets, and inject inlets of the third set of inject inlets that are away from the loading port, while blocking flow of the gas mixture into inject inlets of the third set of inject inlets that are closer to the loading port.

Abstract: A composite material component manufacturing method including a first thermoforming step for creating a first three-dimensional prepreg sheet by thermoforming a thermoplastic first prepreg sheet into a three dimensional shape, a laminate body creating step for creating a prepreg sheet laminate body by layering the first three-dimensional prepreg sheet and a second prepreg sheet; and a laminate body molding step for molding the prepreg sheet laminate body by applying heat and a pressing force to the prepreg sheet laminate body with a pressing device.

Abstract: Provided are a feature transformation device and others enabling feature transformation with high precision.

Abstract: In one embodiment, a substrate support assembly includes a susceptor for supporting a substrate, and a supporting transfer mechanism coupled to the susceptor, the supporting transfer mechanism having a surface for supporting a peripheral edge of the substrate, the supporting transfer mechanism being movable relative to an upper surface of the susceptor.

Abstract: A substrate processing system is disclosed which includes a substrate input/output chamber coupled to a transfer chamber, and one or more processing chambers coupled to the transfer chamber, wherein the substrate input/output chamber includes a plurality of stacked carrier holders, and one of the carrier holders includes a substrate carrier to support a substrate thereon.

Abstract: An air intake device includes: a valve body which includes a rotating shaft; a bearing member which rotatably supports the rotating shaft of the valve body; and an air intake port which includes a concave portion on which the bearing member is disposed, wherein the bearing member includes a positioning portion for positioning the bearing member with respect to the concave portion of the air intake port, facing surfaces which extend from the positioning portion along the concave portion of the air intake port and face each other in an inward radial direction of the rotating shaft with respect to the concave portion, with a gap therebetween, and protruding portions which protrude toward the concave portion of the air intake port from the facing surfaces and seal the gap.

Abstract: An intake device includes: a rotary shaft; and a resin-made intake valve including a rotary shaft press-fit guide hole provided at an end portion in a rotary axis direction to guide the rotary shaft when the rotary shaft is press-fitted, a rotary shaft press-fit hole provided on a more inner side than the rotary shaft press-fit guide hole in the rotary axis direction, the rotary shaft inserted through the rotary shaft press-fit guide hole being press-fitted into the rotary shaft press-fit hole, and a window portion exposing an outer surface of the rotary shaft, the resin intake valve being rotated together with rotation of the rotary shaft in the intake port to change intake air, in which the rotary shaft press-fit guide hole has a tapered shape portion which gradually decreases in cross-sectional area toward the rotary shaft press-fit hole when viewed from the rotary axis direction.

Abstract: Provided is a search system which is configured to search for a registered vector being similar to an input vector among a plurality of registered vectors, on the basis of a degree of similarity between an input vector and a registered vector. The search system includes a partial similarity calculation unit that calculates a degree of partial similarity which is the degree of similarity concerning some of one or more dimensions of the input vector and the registered vector, a limit calculation unit that calculates, on the basis of the degree of partial similarity, an upper limit of the degree of similarity that is expected when the degree of similarity is calculated, and a rejection decision unit that decides, on the basis of the upper limit of the degree of similarity, whether or not to reject the registered vector from a candidate for a search result.

Abstract: An intake device includes: an intake port; and a valve body which is disposed in the intake port and is rotated around a rotation axial line between an open/closed position, in which the valve body includes a valve body main body made of a resin, an elastically deformable seal portion formed to extend along an outer circumferential portion of the valve body main body, and a projection which is provided at a surface part adjacent to a region in which the seal portion is formed in the valve body main body, and has a shape that gradually becomes tapered toward a tip end side.

Abstract: Methods and apparatus for determining an endpoint of a process chamber cleaning process are provided. In some embodiments, a method of monitoring a process being performed in a process chamber includes: performing a cleaning process in a process chamber to remove material deposited on one or more internal surfaces of the process chamber resultant from processes performed within the process chamber; shining a light on a first internal surface being cleaned; detecting the light reflected off of the first internal surface; and terminating the cleaning process based upon the detected light.

Abstract: Provided is a search system which is configured to search for a registered vector being similar to an input vector among a plurality of registered vectors, on the basis of a degree of similarity between an input vector and a registered vector. The search system includes a partial similarity calculation unit that calculates a degree of partial similarity which is the degree of similarity concerning some of one or more dimensions of the input vector and the registered vector, a limit calculation unit that calculates, on the basis of the degree of partial similarity, an upper limit of the degree of similarity that is expected when the degree of similarity is calculated, and a rejection decision unit that decides, on the basis of the upper limit of the degree of similarity, whether or not to reject the registered vector from a candidate for a search result.

Abstract: Provided is a search system which is configured to search for a registered vector being similar to an input vector among a plurality of registered vectors, on the basis of a degree of similarity between an input vector and a registered vector. The search system includes a partial similarity calculation unit that calculates a degree of partial similarity which is the degree of similarity concerning some of one or more dimensions of the input vector and the registered vector, a limit calculation unit that calculates, on the basis of the degree of partial similarity, an upper limit of the degree of similarity that is expected when the degree of similarity is calculated, and a rejection decision unit that decides, on the basis of the upper limit of the degree of similarity, whether or not to reject the registered vector from a candidate for a search result.

Abstract: A neural network capturing a minute pattern variation useful for recognition while maintaining robustness against a pattern variation unrelated to recognition is learned. A preprocessing unit performs, on a set of patterns being to be learned and including a specific pattern variation, a plurality of preprocesses causing different degrees of the specific pattern variation. A network structure determination unit determines, for each of the plurality of preprocesses, a network structure of a neural network having robustness according to a degree of the specific pattern variation after the preprocess. A network learning unit learns, for each of the plurality of preprocesses, the neural network with the network structure associated with the preprocess using the set of patterns after the preprocess.

Abstract: Provided is a search system which is configured to search for a registered vector being similar to an input vector among a plurality of registered vectors, on the basis of a degree of similarity between an input vector and a registered vector. The search system includes a partial similarity calculation unit that calculates a degree of partial similarity which is the degree of similarity concerning some of one or more dimensions of the input vector and the registered vector, a limit calculation unit that calculates, on the basis of the degree of partial similarity, an upper limit of the degree of similarity that is expected when the degree of similarity is calculated, and a rejection decision unit that decides, on the basis of the upper limit of the degree of similarity, whether or not to reject the registered vector from a candidate for a search result.