Abstract: A method for detecting a human behavior includes: obtaining an image to be detected; obtaining a plurality of key points and a plurality of pieces of position information respectively corresponding to the plurality of key points by key-point recognition on the image to be detected; grouping the plurality of key points based on the plurality of pieces of position information to obtain a plurality of key-point groups, the plurality of key-point groups at least including a part of the plurality of key points; and determining a target human behavior based on key points in the plurality of key-point groups.

Abstract: A battery pack and an electric vehicle are provided. The battery pack includes a housing; and a plurality of cells, provided in the housing, wherein the sum V1 of the volumes of the plurality of cells and the volume V2 of the battery pack satisfy V1/V2?55%. The battery pack has a first direction and a second direction perpendicular to each other. A length direction of the cell is arranged along the first direction of the battery pack, and the plurality of cells are arranged along the second direction of the battery pack. The cell comprises a cell body, and the length of the cell body is 400-2500 mm.

Abstract: The invention discloses a synthesizing method of benzimidazole derivant and intermediate, which is characterized by the following: proceeding arylation among aryl halide and primary amine; adopting L-pro as additive, cuprous iodide as catalyst; proceeding molecular arylation coupling reaction for aryl iodide or aryl bromide and primary amine; adopting iodo phenylamine compound and bromophenylamine compound as starting material; synthesizing the material through several-step transformation.

Abstract: The present application relates to the technical field of pharmaceutical chemistry, and particularly relates to compounds with ALK inhibitory activity and preparation method and use thereof. The compounds with ALK inhibitory activity provided by the present application are ALK inhibitors for treating diseases responsive to the inhibition of ALK kinase, can be used for treating ALK-positive diseases, such as tumors and cancers, and have broad application prospects.

Abstract: A system predicts metadata attributes associated with documents using machine learning models. The document may represent an interaction between entities. The system trains machine learning models to predict scores indicating whether a token or a sequence of token of a document represents a metadata attribute. The metadata prediction is used to annotate the document and display to users. The system receives user feedback via the user interface and uses the user feedback to evaluate or retrain the model. The system generates training data by receiving a set of annotated documents and comparing the annotated documents against other documents to identify matching documents. The system determines when to execute the machine learning based metadata prediction based on steps of document workflow executed by the system.

Abstract: A system predicts metadata attributes associated with documents using machine learning models. The document may represent an interaction between entities. The system trains machine learning models to predict scores indicating whether a token or a sequence of token of a document represents a metadata attribute. The metadata prediction is used to annotate the document and display to users. The system receives user feedback via the user interface and uses the user feedback to evaluate or retrain the model. The system generates training data by receiving a set of annotated documents and comparing the annotated documents against other documents to identify matching documents. The system determines when to execute the machine learning based metadata prediction based on steps of document workflow executed by the system.

Abstract: A system predicts metadata attributes associated with documents using machine learning models. The document may represent an interaction between entities. The system trains machine learning models to predict scores indicating whether a token or a sequence of token of a document represents a metadata attribute. The metadata prediction is used to annotate the document and display to users. The system receives user feedback via the user interface and uses the user feedback to evaluate or retrain the model. The system generates training data by receiving a set of annotated documents and comparing the annotated documents against other documents to identify matching documents. The system determines when to execute the machine learning based metadata prediction based on steps of document workflow executed by the system.

Abstract: A configuration for fallback condition from TRP-specific BFR to cell-specific BFR. The apparatus receives a configuration for a cell-specific BFR procedure and a TRP-specific BFR procedure. The apparatus detects a first beam failure at a first TRP of a cell. The apparatus detects a second beam failure at a second TRP of the cell. The apparatus initiates the cell-specific BFR procedure based at least on the first beam failure at the first TRP or the second beam failure at the second BFR.

Abstract: An electronic device includes a carrier having at least one bonding pad, a plurality of electronic elements disposed on the carrier and one of the electronic elements including a substrate and at least one connecting terminal disposed between the substrate and the carrier. The electronic elements are electrically connected to the at least one bonding pad via the at least one connecting terminal.

Abstract: A modulation method and a modulation device for modulating beamforming are provided. The modulation method includes: providing a memory element that stores a beamforming table; providing a driving element; providing a plurality of modulation element groups; providing a signal source to provide at least one initial electromagnetic wave to the modulation element groups; providing a plurality of driving signals corresponding to the modulation element groups according to the beamforming table by the driving element; modulating the at least one initial electromagnetic wave to be a main wave beam with a specific radiation waveform according to the drive signals by the modulation element groups.

Abstract: A fastener plate for use in coupling an underlayment to a roof surface is disclosed. In use, the fastener plate is arranged and configured to enable subsequently applied liquid coatings to flow through a top surface of the fastener plate and into one or more cavities so that the liquid coating seals any openings created by introduction of the fastener. In one embodiment, the fastener plate may include a top surface, a bottom surface, a fastener opening for receiving a fastener, a plurality of openings formed in the top surface, and one or more cavities positioned between the top surface and the bottom surface, the one or more cavities being in fluid communication with the plurality of openings so that the liquid coating can flow through the plurality of openings formed in the top surface and into the one or more cavities to seal any voids created by the fastener.

Abstract: A self-sealing fastener for installation of roofing materials for creating a substantially watertight seal around fastener penetrations where roofing materials or panels are affixed to an underlying roof deck by the fasteners. Also disclosed is a method for installing the self-sealing fastener for attachment of roofing materials to an underlying roof deck with the penetrations areas or holes created by the fasteners being substantially sealed against water penetration.

Abstract: An applicator head for applying a product including a cosmetic, care, or pharmaceutical product onto the keratinous substrate. The applicator head includes an applying member elongated along a central longitudinal axis. The applying member includes a front face, a rear face opposing the front face, a left face, and a right face opposing the left face. The applying member has an axial zigzag groove extending along a length of the applying member on at least one of the front face and the rear face of the applying member. The axial zigzag groove has vertices on both side of the central longitudinal axis of the applying member. Further, the applying member includes side grooves disposed on each side of the axial zigzag groove. The axial zigzag groove and the side grooves outline beads on an outer surface of the applying member.

Abstract: A computer-implemented method for providing merchant recommendations comprises receiving, by a processor, raw merchant embeddings generated from payment transaction records, wherein the raw merchant embeddings include a plurality of embedded features entangled in an embedding space. The processor uses transaction metadata associated with the payment transaction records to determine a subspace of an identified feature within the embedding space. A linear transformation process then removes the subspace of the identified feature from the embedding space to create modified merchant embeddings that are merged and aligned with other ones of the plurality of features within the embedding space. The processor automatically generates a list of merchant rankings based on the modified merchant embeddings, past preferences of a target user using raw user embeddings, and a target region, and provides the list of merchant rankings to the target user.

Abstract: Systems, methods, and computer program products may compare machine learning models by identifying data instances with disagreed predictions and learning from the disagreement. Based on a model interpretation technique, differences between the compared machine learning models may be interpreted. Multiple metrics to prioritize meta-features from different perspectives may also be provided.

Abstract: A method, system, and computer program product is provided for embedding compression and reconstruction. The method includes receiving embedding vector data comprising a plurality of embedding vectors. A beta-variational autoencoder is trained based on the embedding vector data and a loss equation. The method includes determining a respective entropy of a respective mean and a respective variance of each respective dimension of a plurality of dimensions. A first subset of the plurality of dimensions is determined based on the respective entropy of the respective mean and the respective variance for each respective dimension of the plurality of dimensions. A second subset of the plurality of dimensions is discarded based on the respective entropy of the respective mean and the respective variance for each respective dimension of the plurality of dimensions. The method includes generating a compressed representation of the embedding vector data based on the first subset of dimensions.

Abstract: Described are a system, method, and computer program product for debiasing embedding vectors of machine learning models. The method includes receiving embedding vectors and generating two clusters thereof. The method includes determining a first mean vector of the first cluster and a second mean vector of the second cluster. The method includes determining a bias associated with each of a plurality of first candidate vectors and replacing the first mean vector with a first candidate vector based on the bias. The method includes determining a bias associated with each of a plurality of second candidate vectors and replacing the second mean vector with a second candidate vector based on the bias. The method includes repeatedly replacing the first and second mean vectors until an extremum of the bias score is reached, and debiasing the embedding vectors by linear projection using a direction defined by the first and second mean vectors.

Abstract: A manufacturing method of an electronic device includes the following steps. First, a curved substrate is formed. Next, a first conductive layer is formed on a first insulating layer. Then, a second conductive layer is formed on a second insulating layer. Thereafter, the first insulating layer and the second insulating layer are bonded to the curved substrate. The manufacturing method of the electronic device may reduce problems such as peeling and/or offset of the manufactured electronic device.

Abstract: Provided is a system for analyzing features associated with entities using an embedding tree, the system including at least one processor programmed or configured to receive a dataset associated with a plurality of entities, wherein the dataset comprises a plurality of data instances for a plurality of entities. The processor may be programmed or configured to generate at least two embeddings based on the dataset and determine split criteria for partitioning an embedding space of at least one embedding tree associated with the dataset based on feature data associated with an entity and embedding data associated with the at least two embeddings. The processor may be programmed or configured to generate at least one embedding tree having a plurality of nodes based on the split criteria. Methods and computer program products are also provided.

Abstract: Curable polyolefin composition comprising: (A) polyolefin having at least two aliphatic unsaturated bonds per molecule; (B) wax with melting point of 30 to 100° C.; (C) organopolysiloxane having at least two silicon atom-bonded hydrogen atoms per molecule; and (D) catalytic amount of hydrosilylation reaction catalyst, wherein content of component (A) is 20 to 80 mass %, content of component (B) is 10 to 75 mass %, and content of component (C) is 1 to 20 mass %, each based on the total mass of components (A) to (D). The composition can be cured to form a cured product capable of storing and releasing thermal energy, and preventing the wax leaking/pumping out during heat cycling.