Abstract: A non-transitory computer-readable storage medium storing an information processing program that causes at least one computer to execute a process, the process includes acquiring an update amount of a classification criterion of a classification model in retraining, the classification model being trained by using a first dataset, the classification model classifying input data into one of a plurality of classes, the retraining being performed by using a second dataset; and detecting data with a largest change amount among the second dataset when changing each piece of data included in the second dataset so as to decrease the update amount.

Abstract: An information processing method comprising: for a classification model for classifying input data into one or another of plural classes that was trained using a first data set, identifying, in a second data set that is different from the first data set one or more items of data having a specific datum of which a degree of contribution to a change in a classification criterion is greater than a predetermined threshold, the classification criterion being a classification criterion of the classification model during re-training based on the second data set; and, from among the one or more items of data, detecting an item of data, for which a loss reduces for the classification model by change to the classification criterion by re-training based on the second data set, as an item of data of an unknown class not contained in the plural classes.

Abstract: A non-transitory computer-readable recording medium stores a determination program for causing a computer to execute processing including: re-training a classification model that has been trained by using a first data set and that classifies input data into any one of a plurality of classes by using a loss calculatable based on a second data set that is different from the first data set; and determining, in a case where a change in a classification standard of the classification model based on the loss is a predetermined standard or more before and after re-training, that unknown data that is not classified into any one of the plurality of classes is included in the second data set.

Abstract: An information processing device acquires output image data that is acquired by inputting image data indicating a pseudo-shadow area to an auto-encoder that is generated by machine learning using label image data contained in training data, the label image data indicating a shadow area in ultrasound image data of a captured target. The information processing device generates augmented data corresponding to the training data by combining the acquired output image data with the ultrasound image data.

Abstract: A non-transitory computer-readable recording medium stores a generation program for causing a computer to execute a process including: with data included in each of a plurality of data sets, training a feature space in which a distance between pieces of the data included in a same domain is shorter and the distance of the data between different domains is longer; and generating labeled data sets by integrating labeled data included within a predetermined range in the trained feature space, among a plurality of pieces of the labeled data.

Abstract: Provided is a steel sheet comprising a sheet thickness center part and a first surface layer softened part and a second surface layer softened part respectively arranged at two sides of the sheet thickness center part, wherein the first surface layer softened part and second surface layer softened part have 10 ?m or more average thicknesses and have average Vickers hardnesses of 0.90 time or less of the average Vickers hardness of a sheet thickness ½ position, and the first surface layer softened part has an average Vickers hardness of 1.05 times or more the average Vickers hardness of the second surface layer softened part.

Abstract: Provided is a steel sheet comprising a sheet thickness center part and a first surface layer part and a second surface layer part respectively arranged at two sides of the sheet thickness center part, wherein the first surface layer part and second surface layer part respectively independently have thicknesses of more than 10 ?m to 30% or less of the sheet thickness, the first surface layer part and second surface layer part have average Vickers hardnesses different from the average Vickers hardness of a sheet thickness ½ position, and a first hardness cumulative value at a region from a surface of the first surface layer part side to 30% of the sheet thickness is 1.05 times or more of a second hardness cumulative value at a region from a surface of the second surface layer part side to 30% of the sheet thickness.

Abstract: A driver including: an ejection part to which a fastener is supplied; a driver blade which moves from a first position toward a second position and drives the fastener into a driven member; and a rack provided to the driver blade. The driver further includes: a rotary component engaging with the rack and moving the driver blade from the second position to the first position; and a lock plate engaging with the rack. The driver blade moves from the second position to the first position while the rotary component rotates once, the rotary component is released from engaging with the rack after the driver blade moves from the second position to the first position, and moves from the first position to the second position, and the lock plate is engageable with the rack when the driver blade stops before reaching the second position from the first position.

Abstract: A program for causing a computer to execute processing including: acquiring a plurality of datasets, each of which includes data values associated with a label, the data values having properties different for each dataset; calculating an index indicating a degree of a difference between first and second datasets by using a data value in the second dataset; calculating accuracy of a prediction result for the second dataset, predicted by a prediction model trained using the first dataset; specifying a relationship between the index and the accuracy of the prediction result from the prediction model, based on the index and the accuracy calculated for each of a plurality of combinations of the first and second datasets; and estimating accuracy of the prediction result from the prediction model for a third dataset including data values without labels based on the specified relationship and the index between the first and third datasets.

Abstract: A learning method executed by a computer, the learning method including augmenting original training data based on non-stored target information included in the original training data to generate a plurality of augmented training data, generating a plurality of intermediate feature values by inputting the plurality of augmented training data to a learning model, and learning a parameter of the learning model such that, with regard to the plurality of intermediate feature values, each of the plurality of intermediate feature values generated from a plurality of augmented training data, augmented from reference training data, becomes similar to a reference feature value.

Abstract: This hot-rolled steel sheet has a predetermined chemical composition and predetermined metallographic structure, a ratio between a maximum depth of a region where, on one surface, a rotation angle between a normal line of the one surface and a (011) pole near the normal line of the one surface is 5° or less and a maximum depth of a region where, on the other surface, a rotation angle between a normal line of the other surface and a (011) pole near the normal line of the other surface is 5° or less is 1.00 to 1.20, and a tensile strength is 1150 MPa or more.

Abstract: Adopted are a hot-rolled steel sheet having a predetermined chemical composition, in which a region, where a rotation angle between a normal line of a surface and a (011) pole near the normal line becomes 5° or less, is 0.150 or less from the surface in terms of a sheet thickness direction position standardized by a sheet thickness, and a region, where the rotation angle becomes 20° or more, is 0.250 or more from the surface in terms of the sheet thickness direction position standardized by the sheet thickness and a manufacturing method thereof.

Abstract: According to one aspect of the present invention, there is provided a T-joint including a first steel sheet, a second steel sheet, and a fillet welded part, in which the sheet thickness of the second steel sheet is 6.0 mm or less, the second steel sheet is stood on a first surface of the first steel sheet, the fillet welded part joins the first surface of the first steel sheet and a first surface of the second steel sheet to each other, at least one of the first surface of the first steel sheet or the first surface of the second steel sheet includes a zinc-based plating, an abutting end portion of the second steel sheet on a second surface side of the second steel sheet has an inclined surface, and in a cross section taken along a sheet thickness direction of the first steel sheet and a sheet thickness direction of the second steel sheet, the inclined surface forms an acute angle with respect to the first surface of the first steel sheet.

Abstract: There is provided a cutting method for cutting a workpiece using a cutting tool comprising a die and a punch, including: arranging the workpiece between the die and the punch, and in a state in which a wedge-shaped first cutting part of the die and a wedge-shaped second cutting part of the punch are opposed, pushing the punch relatively to the die side to cut the workpiece; wherein: a front end angle ?1 of the first cutting part and a front end angle ?2 of the second cutting part are each 100 or more and 1200 or less, and a front end radius R1 of the first cutting part and a front end radius R2 of the second cutting part are each 0.5% or more and 35.0% or less of a sheet thickness.

Abstract: The present invention has as its technical problem to provide a punching method able to stably secure a stretch-flangeability equal to or greater than the case of using a punch having a flat blade at its bottom surface. The punching method of the present invention is a method of punching a metal sheet using a punching and shearing device provided with a punch having an upper blade having a horizontal part with respect to the cutting line in part and having parts other than the horizontal part comprised of inclined parts and a die having a lower blade wherein the upper blade used for the punching process has a shape where the inclined parts first contact the metal sheet at the time of a punching process.

Abstract: A blank (100) includes a main portion (110) that is made of steel having a tensile strength of 1450 MPa or more and a softened portion (120), the ratio of Vickers hardness of the softened portion (120) to the Vickers hardness of the main portion (110) is 0.7 or more and 0.95 or less, and the softened portion (120) is disposed at a position different from a position of the main portion (110) in an in-plane direction. A structural member (200) includes a first member (210), a second member (220), and a weld (W) at which the first member (210) and the second member (220) are welded to each other.

Abstract: This steel sheet is a steel sheet (100) formed by causing end surfaces of a first sheet material (111) and a second sheet material (113) to abut each other in an in-plane direction and welding the first sheet material (111) and the second sheet material (113) via a strip-shaped welded part (115), and in which a softened part (120) that is softened more than other parts in the welded part (115) is formed in at least a part of the welded part (115), and on a first end surface of the steel sheet in which an end part of the welded part (115) in a longitudinal direction is formed, a region in which the softened part (120) is not formed is provided in at least a part of the end part of the welded part (115) in the longitudinal direction, and a maximum value of a depth of the softened part (120) in a sheet thickness direction is, as a ratio to a sheet thickness of the steel sheet (100), 50% or less.

Abstract: A cutting method using a stamping press according to the present disclosure is a method to cut a workpiece that is configured from a first metal sheet and a second metal sheet joined at a weld portion and that has a heat-affected zone around the weld portion, in which the workpiece is cut using a punch. The punch includes a flat portion and a projecting portion projected more toward the workpiece than a flat portion of the punch. The workpiece is positioned with respect to the punch at a position such that the projecting portion starts cutting at least at one out of the heat-affected zone or the weld portion before the flat portion cuts the workpiece. The workpiece is then cut by moving the punch and a die relative to each other in this state of positioning so as to shear across the weld portion on the workpiece.

Abstract: There is provided a hole widening method including a preparing process of preparing a forming tool which has a diameter-increasing portion increasing in diameter from a front end side toward a rear end side and a line-shaped projection formed to protrude outward from a surface of the diameter-increasing portion, and a workpiece in which a pilot hole is formed; and a hole widening process of successively widening the pilot hole by pushing the forming tool into the pilot hole such that the line-shaped projection of the forming tool comes into point contact with a part of a circumferential edge portion of the pilot hole in the workpiece two times or more, and forming a stretched flange.

Abstract: In a breaking prediction method of predicting a breaking portion of a component, which is obtained by forming a metal sheet, by using a finite element method, the breaking portion is easily and reliably extracted.