Abstract: A representation learning apparatus executing: calculating a latent vector Sx in a latent space of the target data x using a first model parameter, calculate a non-interest latent vector Zx in a latent space of an non-interest feature included in the target data x and a non-interest latent vector Zb in the latent space of a non-interest data using a second model parameter, calculate a similarity S1 obtained by correcting a similarity between the latent vector Sx and its representative value S?x by a similarity between the latent vector Zx and its representative value Z?x, and a similarity S2 between the latent vector Zb and its representative value Z?b, and update the first and/or the second model parameter based on the loss function including the similarity S1 and S2.

Abstract: According to one embodiment, a data labeling work support apparatus includes a processor including hardware. The processor acquires a first label assigned to data. The processor acquires the data. The processor extracts a feature of the data. The processor groups the data based on a similarity or a distance of the feature. The processor assigns a second label to the grouped data. The processor calculates a degree of matching between the first label and the second label. The processor outputs information regarding a combination of the first label and the second label having a low degree of matching.

Abstract: According to an embodiment, a learning device includes a memory and one or more processors coupled to the memory. The one or more processors are configured to: generate a transformation matrix from learning data in which feature quantities and target values are held in a corresponding manner; and learn about parameters of a neural network which includes nodes equal in number to the number of rows of the transformation matrix, a first output layer representing first estimation distribution according to the values of the nodes, and a second output layer representing second estimation distribution decided according to the product of the transformation matrix and the first estimation distribution.

Abstract: According to one embodiment, an information processing apparatus includes a processing circuit. The processing circuit calculates a first input/output error related to normal data and a second input/output error related to pseudo abnormal data different from the normal data, for each of a plurality of autoencoders having different network structures. The processing circuit outputs relational data indicating a relation between the network structure and the first input/output error and the second input/output error.

Abstract: According to one embodiment, an information processing apparatus includes a processor. The processor generates a machine learning model by coupling one feature extractor to each of a plurality of predictors, the feature extractor being configured to extract a feature amount of data. The processor trains the machine learning model for a specific task using a result of ensembling a plurality of outputs from the predictors.

Abstract: An actuator capable of attaining high output. The actuator includes a frame structure part that forms a frame structure surrounding a housing part, and a volume change part housed in the housing part. The volume change part increases a volume thereof by input of external energy. The frame structure part has a higher Young's modulus than a Young's modulus of the volume change part. The housing part has an anisotropic shape, with a maximum width in first direction of the housing part longer than a maximum width in second direction different from the first direction of the housing part.

Abstract: According to one embodiment, an information processing apparatus includes a processing circuit. The processing circuit calculates a first input/output error related to normal data and a second input/output error related to pseudo abnormal data different from the normal data, for each of a plurality of autoencoders having different network structures. The processing circuit outputs relational data indicating a relation between the network structure and the first input/output error and the second input/output error.

Abstract: The present invention relates to a separation membrane element including a supply-side channel member, in which: the supply-side channel member has a net shape in which plural fibrous rows X including fibrous objects A and plural fibrous rows Y including fibrous objects B cross each other sterically to form intersections; at least one of the fibrous objects A and the fibrous objects B have a large diameter portion and a small diameter portion along a longitudinal direction; at least one of the fibrous objects A and the fibrous objects B include a thread that is thinner at a central portion located between intersection portions than at the large diameter portion; and a fiber between an arbitrary intersection and an adjacent intersection is a tapered fiber whose diameter increases like a taper in a direction from one intersection to the other intersection.

Abstract: According to one embodiment, the importance analysis apparatus includes an importance calculator and a distribution calculator. Based on a trained model and a plurality of input data samples, the importance calculator calculates an importance of each of a plurality of feature amounts of each of the input data samples. The distribution calculator calculates a distribution of the importances of each of the feature amounts across the input data samples.

Abstract: An object of the present invention is to provide a separation membrane having high permeability and selective removability of divalent/monovalent ions. The separation membrane of the present invention includes a supporting membrane and a separation functional layer formed on the supporting membrane, in which the separation functional layer contains a polymerized product of a polyfunctional amine with a polyfunctional acid halide, the polyfunctional amine contains a polyfunctional aliphatic amine as a main component, the separation functional layer has a hollow protuberant structure, and the separation functional layer has a relative surface area of 1.1-10.0.

Abstract: An actuator capable of attaining high output. The actuator includes a frame structure part that forms a frame structure surrounding a housing part, and a volume change part housed in the housing part. The volume change part increases a volume thereof by input of external energy. The frame structure part has a higher Young's modulus than a Young's modulus of the volume change part. The housing part has an anisotropic shape, with a maximum width in first direction of the housing part longer than a maximum width in second direction different from the first direction of the housing part.

Abstract: According to an embodiment, a learning device includes a memory and one or more processors coupled to the memory. The one or more processors are configured to: generate a transformation matrix from learning data in which feature quantities and target values are held in a corresponding manner; and learn about parameters of a neural network which includes nodes equal in number to the number of rows of the transformation matrix, a first output layer representing first estimation distribution according to the values of the nodes, and a second output layer representing second estimation distribution decided according to the product of the transformation matrix and the first estimation distribution.

Abstract: According to an embodiment, a learning device includes one or more processors. The processors calculate a latent vector of each of a plurality of first target data, by using a parameter of a learning model configured to output a latent vector indicating a feature of a target data. The processors calculate, for each first target data, first probabilities that the first target data belongs to virtual classes on an assumption that the plurality of first target data belong to the virtual classes different from each other. The processors update the parameter such that a first loss of the first probabilities, and a second loss that is lower as, for each of element classes to which a plurality of elements included in each of the plurality of first target data belong, a relation with another element class is lower, become lower.

Abstract: According to one embodiment, a communication device includes one or more processors. The processors decide on a first-type parameter representing a conversion operation for converting input data into first-type conversion data. The processors calculate first-type predicted distributions based on second-type parameters, each representing one of the plurality of clusters. The processors update the first-type parameter and the second-type parameters so as to achieve optimization of first-type differences representing differences between the first-type predicted distributions and a target distribution, and second-type differences representing differences between the first-type predicted distributions and second-type predicted distributions that indicate probability at which second-type converted data. The second-type converted data is obtained by converting data, which is formed by augmentation of the input data, using the first-type parameter belongs to the clusters.

Abstract: The present invention relates to a separation membrane element including: a separation membrane; and a permeate-side channel material disposed on a permeate side of the separation membrane, wherein the permeate-side channel material is a rugged sheet object having a recess and a protrusion on at least one face thereof, the rugged sheet object is composed of a porous region formed of a through-hole in a thickness direction of the rugged sheet and a non-porous region other than the porous region, and a rate of a number of the through-hole in the recess of the rugged sheet object to a total number of the through-hole present in the rugged sheet object is 80% or more.

Abstract: According to one embodiment, a communication device includes one or more processors. The processors decide on a first-type parameter representing a conversion operation for converting input data into first-type conversion data. The processors calculate first-type predicted distributions based on second-type parameters, each representing one of the plurality of clusters. The processors update the first-type parameter and the second-type parameters so as to achieve optimization of first-type differences representing differences between the first-type predicted distributions and a target distribution, and second-type differences representing differences between the first-type predicted distributions and second-type predicted distributions that indicate probability at which second-type converted data. The second-type converted data is obtained by converting data, which is formed by augmentation of the input data, using the first-type parameter belongs to the clusters.

Abstract: A polyamide multifilament has a tensile strength at 15% elongation of 4.0 to 6.0 cN/dtex, a strength-elongation product of 10.0 cN/dtex or more, and a yarn unevenness (U %) of 1.2 or less. The high-strength polyamide multifilament makes it possible to obtain stockings having high softness, durability, and transparency, and a lace knit in which patterns have a beautiful appearance.

Abstract: The present invention relates to a separation membrane element including: a separation membrane; and a permeate-side channel material disposed on a permeate side of the separation membrane, wherein the permeate-side channel material is a rugged sheet object having a recess and a protrusion on at least one face thereof, the rugged sheet object is composed of a porous region formed of a through-hole in a thickness direction of the rugged sheet and a non-porous region other than the porous region, and a rate of a number of the through-hole in the recess of the rugged sheet object to a total number of the through-hole present in the rugged sheet object is 80% or more.

Abstract: According to an embodiment, a learning device includes a calculator and a learner. The calculator is configured to calculate a value of a first objective function and a value of a second objective function. The first objective function includes smoothness that indicates smoothness of a local distribution of an output of a model, and is used to estimate a first model parameter for determining the model. The second objective function is used to estimate, with a second model parameter that is a hyperparameter of a learning method of learning, the model by using the first objective function. The second model parameter to be estimated is closer to a distance scale of learning data. The learner is configured to update the first model parameter and the second model parameter so that the value of the first objective function and the value of the second objective function are optimized.

Abstract: The present invention provides a separation membrane element packed with a rugged sheet object, the separation membrane element being effective in attaining both stabilization of steps for producing the separation membrane element and an increase in fresh-water production rate. The present invention relates to a separation membrane element including: a separation membrane; and a permeate-side channel material disposed on a permeate side of the separation membrane, in which the permeate-side channel material is a porous sheet object having a recess and a protrusion on at least one face thereof, the recess being a coarsely porous region and the protrusion being a densely porous region.