Abstract: A self-supervised representation learning method includes: outputting, using one of two neural networks, a first parameter that is a parameter of a probability distribution from one of two items of image data obtained by applying data augmentation to one training image obtained from training data; outputting a second parameter that is a parameter of a probability distribution from an other one of the two items of image data, using an other one of the two neural networks; and training the two neural network to optimize an objective function for bringing the two items of image data close to each other, the objective function including a likelihood of the probability distribution of the second parameter.

Abstract: A substrate holder includes a central support portion configured to support a central portion of a substrate, and an circumferential support portion arranged on an outside of the central support portion and configured to support a circumferential portion on an outside of the central portion, and the circumferential support portion is configured to support the circumferential portion so that at least a partial region of the circumferential portion is curved toward the substrate holder with a curvature greater than that of the central portion.

Abstract: A semiconductor device includes a gate electrode on a substrate, a gate insulating film on the gate electrode, an oxide semiconductor film via the gate insulating film on the gate electrode, a source electrode and a drain electrode on the oxide semiconductor film, a protective film provided on the source electrode and the drain electrode; and a conductive layer provided on the protective film and overlapped on the oxide semiconductor layer. The protective film includes a first silicon oxide film and a first silicon nitride film. The first oxide film is in contact with the oxide semiconductor layer. The gate insulating film includes a second silicon nitride film and a second silicon oxide film. The second silicon oxide film is in contact with the oxide semiconductor layer. The oxide semiconductor layer has a first region located between the source electrode and the drain electrode in a plan view.

Abstract: An information processing apparatus performs face detection processing to detect a face area and the orientation of a face from an image captured by an imaging unit, brightness reduction processing in which, when the orientation of the face detected by the face detection processing changes from a first orientation to a second orientation while the screen brightness of a display unit is controlled to a first screen brightness, the screen brightness is reduced from the first screen brightness to a second screen brightness after first time has elapsed, and imaging state determination processing to determine whether an imaging state is a first imaging state where the imaging unit can capture an image in an image brightness capable of detecting the face area or a second imaging state where the imaging unit cannot capture any image in the image brightness capable of detecting the face area.

Abstract: An information processing apparatus performs face detection processing to detect a face area with a face captured therein from an image captured by an imaging unit, brightness reduction processing in which, when the face area is no longer detected from a state where the face area is detected by the face detection processing while the screen brightness of a display unit is controlled to a first screen brightness, the screen brightness is reduced from the first screen brightness to a second screen brightness after first time has elapsed, and imaging state determination processing to determine whether an imaging state is a first imaging state where the imaging unit can capture an image in an image brightness capable of detecting the face area or a second imaging state where the imaging unit cannot capture any image in the image brightness capable of detecting the face area.

Abstract: Provided are a regenerated-styrene-based resin composition that has a high rate of recycled material content, excellent impact resistance, excellent creep resistance in a thin-walled thickness, as well as a safeness in that the flammability complies with UL94HB, and a molded object including the resin composition. In a regenerated-styrene-based resin composition including a styrene-based recycled material (A) and a styrene-based virgin material (B), the content of the styrene-based recycled material (A) being 40% by mass or more and 90% by mass or less relative to 100% by mass of (A) and (B) in total, the content of a rubber-like polymer is 5.0% by mass or more and 7.0% by mass or less, the content of toluene insoluble matter is 21% by mass or less, and a bending strength, determined in accordance with JIS K7171, is 47 MPa or more.

Abstract: An information processing device for controlling a control target device to be controlled includes: a storage unit configured to store layout constraint condition information including information indicating relative positions of a first object disposed in a space in which the control target device is positioned and a second object different from the first object disposed in the space; and a control unit configured to generate map information indicating a map in the space based on a position of the control target device, at least one of first object-device relative position information indicating relative positions of the first object and the control target device or second object-device relative position information indicating relative positions of the second object and the control target device, and the layout constraint condition information stored in the storage unit.

Abstract: A substrate processing device reduces a surface of a substrate. The surface includes a metal layer. The substrate processing device includes a chamber body, a hot plate, a plasma supply unit, and a controller. The hot plate is accommodated in the chamber body and configured to set the substrate. The plasma supply unit is configured to supply plasma of a hydrogen gas to the chamber body. The controller is configured to execute a degassing process and a reducing process. In the degassing process, the controller drives the hot plate to remove an adsorbate from the surface before driving the plasma supply unit. In the reducing process, the controller drives the plasma supply unit after driving the hot plate to supply the plasma to the surface that has undergone the degassing process.

Abstract: A learning method for learning an action of an agent using model-based reinforcement learning is provided. The learning method includes: obtaining time series data indicating states and actions of the agent when the agent performs a series of actions; establishing a dynamics model by performing supervised learning using the time series data obtained; deriving a plurality of candidates for an action sequence of the agent from variational inference using a mixture model as a variational distribution, based on the dynamics model; and outputting, as the action sequence of the agent, one candidate selected from among the plurality of candidates derived.

Abstract: An object tracking method includes: predicting a prediction distribution which is a distribution of each of prediction pose locations of first objects in a current frame by inputting pose time-sequence data to a pose predictor; estimating a set of locations of each of second objects in the current frame by inputting the current frame to a pose estimator; obtaining identification information indicating a one-to-one correspondence between the second objects and the first objects and location information of the second objects by matching the set of locations and the prediction distribution, to output the obtained identification information and location information; and updating the pose time-sequence data by adding, to the pose time-sequence data and based on the obtained identification information and location information, data which includes coordinates indicating a pose of each of the first objects and is obtained from the predicted prediction distribution.

Abstract: A substrate holder includes a central support portion configured to support a central portion of a substrate, and an circumferential support portion arranged on an outside of the central support portion and configured to support a circumferential portion on an outside of the central portion, and the circumferential support portion is configured to support the circumferential portion so that at least a partial region of the circumferential portion is curved toward the substrate holder with a curvature greater than that of the central portion.

Abstract: An object tracking method includes: predicting a prediction distribution which is a distribution of each of prediction pose locations of first objects in a current frame by inputting pose time-sequence data to a pose predictor; estimating a set of locations of each of second objects in the current frame by inputting the current frame to a pose estimator; obtaining identification information indicating a one-to-one correspondence between the second objects and the first objects and location information of the second objects by matching the set of locations and the prediction distribution, to output the obtained identification information and location information; and updating the pose time-sequence data by adding, to the pose time-sequence data and based on the obtained identification information and location information, data which includes coordinates indicating a pose of each of the first objects and is obtained from the predicted prediction distribution.

Abstract: A learning method for learning an action of an agent using model-based reinforcement learning is provided. The learning method includes: obtaining time series data indicating states and actions of the agent when the agent performs a series of actions; establishing a dynamics model by performing supervised learning using the time series data obtained; deriving a plurality of candidates for an action sequence of the agent from variational inference using a mixture model as a variational distribution, based on the dynamics model; and outputting, as the action sequence of the agent, one candidate selected from among the plurality of candidates derived.

Abstract: A substrate bonding apparatus that bonds a first substrate and a second substrate together, comprising a joining section that joins the first substrate and second substrate together aligned to each other for stacking; a detecting section that detects an uneven state on at least one of the first substrate and second substrate prior to joining by the joining section; and a determining section that determines whether the uneven state detected by the detecting section satisfies a predetermined condition, wherein the joining section does not join the first substrate and the second substrate if it is determined by the determining section that the uneven state does not satisfy the predetermined condition.

Abstract: An unmanned aerial vehicle includes: a photographic-device mounting portion to which a photographic device is mounted; a communicator that communicates with the photographic device; and an authentication processor that performs authentication on the photographic device by using the communicator and that determines whether or not the photographic device is a legitimate device that satisfies a predetermined aerial photography requirement. Upon determining that the photographic device is not a legitimate device, the authentication processor performs operation limitation on a predetermined function of the unmanned aerial vehicle so that the photographic device is incapable of performing aerial photography.

Abstract: An unmanned aerial vehicle includes: a memory that stores first information acquired and second information acquired, the first information indicating the current position of the unmanned aerial vehicle, the second information indicating the current position of a manipulation device that transmits a command to remotely manipulate the unmanned aerial vehicle; a receiver that receives information transmitted from a transmission source, the information including an emergency warning; a processor; and a controller that controls the unmanned aerial vehicle. The processor compares a target region which is indicated in the information and for which the emergency warning is intended and the current position of the unmanned aerial vehicle and compares the target region and the current position of the manipulation device. The controller evacuates the unmanned aerial vehicle in accordance with the results of the comparisons.

Abstract: A control device for performing optimal control by path integral includes a neural network section including a machine-learned dynamics model and cost function, an input section that inputs a current state of a control target and an initial control sequence for the control target into the neural network section, and an output section that outputs a control sequence for controlling the control target, the control sequence being calculated by the neural network section by path integral from the current state and the initial control sequence by using the dynamics model and the cost function. Here, the neural network section includes a second recurrent neural network incorporating a first recurrent neural network including the dynamics model.

Abstract: A detector comprises a plurality of photoelectric converters to output an electrical signal corresponding to an incident light, and a plurality of filter circuits provided corresponding to each of the plurality of photoelectric converters or to each of a plurality of element groups respectively including a predetermined number of the photoelectric converters of the plurality of photoelectric converters, the plurality of filter circuits attenuating a signal having a predetermined frequency from the electrical signal output from the plurality of photoelectric converters, In the above-described detector, the plurality of photoelectric converters may be provided in a first substrate, and the plurality of filter circuits may be provided in a second substrate laminated on the first substrate.

Abstract: A device that supports a performer of a wind instrument, the device including: a processor; and a memory, in which the processor acquires data indicating a myoelectric potential value measured by, a myoelectric sensor arranged on a surface of the face of a performer, the memory stores a table including a listing of conditions for the myoelectric potential value and a corresponding listing of support information for playing the wind instrument, and the processor determines whether the acquired myoelectric potential value satisfies at least a portion of a target condition included in the listing of conditions, determines support information corresponding to the target condition, when the acquired myoelectric potential value is determined to satisfy the at least a portion of the target condition, generates a support image corresponding to the determined support information, and controls a display to display the generated support image.

Abstract: An unmanned aerial vehicle includes: a memory that stores first information acquired and second information acquired, the first information indicating the current position of the unmanned aerial vehicle, the second information indicating the current position of a manipulation device that transmits a command to remotely manipulate the unmanned aerial vehicle; a receiver that receives information transmitted from a transmission source, the information including an emergency warning; a processor; and a controller that controls the unmanned aerial vehicle. The processor compares a target region which is indicated in the information and for which the emergency warning is intended and the current position of the unmanned aerial vehicle and compares the target region and the current position of the manipulation device. The controller evacuates the unmanned aerial vehicle in accordance with the results of the comparisons.