Abstract: Systems and methods for providing deep learning models capable of performing joint representation learning and new category discovery on a mixture of labeled and unlabeled data, which may include single- and multi-modal data. In some examples, a flexible end-to-end framework uses unified contrastive learning on labeled and unlabeled data based on both instance discrimination and category discrimination, and further uses Winner-Take-All hashing to generate a pseudo-label based on the similarity between each pair of unlabeled data points that can be used to train the model to generate clustering assignments for each unlabeled data point. In some examples, the unified contrastive learning may be further based on cross-modal discrimination.

Abstract: The present application provides a dispensing nozzle for use with a flow of a diluent and a flow of a concentrate. The beverage dispenser may include an annular concentrate path of the flow of the concentrate and an annular diluent path surrounding at least in part the annular concentrate path for the flow of the diluent. The annular diluent path may include a shallow angle leading towards the flow of the concentrate such that the flow of the diluent and the flow of the concentrate mix in or downstream of the dispensing nozzle.

Abstract: In one embodiment, a method includes receiving a query inputted by a user; identifying one or more objects matching the query; generating a reconstructed embedding of the query based on one or more term embeddings associated with one or more of the n-grams of the query, respectively; calculating, for each identified object, a relevance-score based on a similarity metric of the reconstructed embedding of the query and an object embedding associated with the identified object; generating one or more search results based on the calculated relevance-scores, each search result corresponding to one of the identified objects; and sending a search-results interface for display, the search-results interface comprising one or more of the search results, each search result comprising a reference to its corresponding identified object.

Abstract: A control interface for controlling the operation of a dispensing device may be provided. The control interface may include a display screen for individually presenting each of multiple options that may be utilized for controlling various operations associated with the dispensing device. The control interface may further include a navigation control, in communication with the display screen, which may be utilized to individually navigate through each of the options on the display screen. The display screen may be actuated to select a displayed option that is associated with performing a dispensing device operation.

Abstract: A dispenser system for providing variable carbonation can include a carbonated water source in communication with a carbonated water valve for controlling flow of carbonated water through the dispenser system; a still water source in communication with a still water valve for controlling flow of still water through the dispenser system; a nozzle in fluid communication with the carbonated water source and still water source; a processing device adapted to modulate the carbonated water and still water valves within a modulate cycle to achieve a ratio of carbonated to non-carbonated water; and a pour mechanism adapted to provide an input to the processing device to dispense the beverage.

Abstract: In one embodiment, a method includes accessing user-concept scores for a first set of users, wherein each user-concept score is associated with a user-concept pair; calculating recommended user-concept scores for a subset of user-concept pairs in a second set of users. The first set of users may be discrete from the second set of users. A recommendation-algorithm may compute the recommended user-concept scores for a user-concept pair by optimizing an objective function comprising a plurality of predicted rating functions. Each predicted rating function may be determined using a user score, a concept score, a user-bias value associated with the user, as well as a concept-bias value associated with the concept. Finally, the method may include sending recommendations for one or more concepts based on the recommended user-concept scores for the second set of users.

Abstract: A dispenser system for providing variable carbonation can include a carbonated water source in communication with a carbonated water valve for controlling flow of carbonated water through the dispenser system; a still water source in communication with a still water valve for controlling flow of still water through the dispenser system; a nozzle in fluid communication with the carbonated water source and still water source; a processing device adapted to modulate the carbonated water and still water valves within a modulate cycle to achieve a ratio of carbonated to non-carbonated water; and a pour mechanism adapted to provide an input to the processing device to dispense the beverage.

Abstract: A beverage dispensing system and method include a beverage dispensing device that has a processor and a memory accessible to the processor. The memory stores instructions that when executed by the processor implement a method that includes receiving a beverage selection, and comparing the beverage selection to predefined controls. The selected beverage is dispensed in response to the comparison.

Abstract: In one embodiment, a set of user groups of a social-networking system may be accessed. A first subset of the user groups may be determined for a particular user of the social-networking system, based on one or more filtering criteria. A number of recommendation-source processes may be applied to the first subset to determine a number of second subsets of the first set. Each recommendation-source process may represent a particular recommendation source. The second subsets may be combined into a list of user groups. The list of user groups may be ranked, and sent to the particular user.

Abstract: Systems, methods, and apparatuses for tracking trends at an online presence are described. In some embodiments, an interactivity data structure and a tenured interactivity data structure are utilized in determining trends quickly.

Abstract: In one embodiment, a method includes receiving a query inputted by a user; identifying one or more objects matching the query; generating a reconstructed embedding of the query based on one or more term embeddings associated with one or more of the n-grams of the query, respectively; calculating, for each identified object, a relevance-score based on a similarity metric of the reconstructed embedding of the query and an object embedding associated with the identified object; generating one or more search results based on the calculated relevance-scores, each search result corresponding to one of the identified objects; and sending a search-results interface for display, the search-results interface comprising one or more of the search results, each search result comprising a reference to its corresponding identified object.

Abstract: The present application provides a dispensing nozzle for use with a flow of a diluent and a flow of a concentrate. The beverage dispenser may include an annular concentrate path of the flow of the concentrate and an annular diluent path surrounding at least in part the annular concentrate path for the flow of the diluent. The annular diluent path may include a shallow angle leading towards the flow of the concentrate such that the flow of the diluent and the flow of the concentrate mix in or downstream of the dispensing nozzle.

Abstract: In one embodiment, a method includes accessing user-concept scores for a first set of users, wherein each user-concept score is associated with a user-concept pair; calculating recommended user-concept scores for a subset of user-concept pairs in a second set of users. The first set of users may be discrete from the second set of users. A recommendation-algorithm may compute the recommended user-concept scores for a user-concept pair by optimizing an objective function comprising a plurality of predicted rating functions. Each predicted rating function may be determined using a user score, a concept score, a user-bias value associated with the user, as well as a concept-bias value associated with the concept. Finally, the method may include sending recommendations for one or more concepts based on the recommended user-concept scores for the second set of users.

Abstract: In one embodiment, a method includes accessing a social graph comprising a plurality of nodes and edges, accessing user-concept scores for a first set of users nodes of the plurality for nodes, respectively, generating a recommendation-algorithm for estimating recommended user-concept scores for all user-concept pairs in the first set of user nodes and the plurality of concept nodes, and calculating recommended user-concept scores for a second set of user nodes of the plurality of nodes.

Abstract: Systems, methods, and apparatuses for tracking trends at an online presence are described. In some embodiments, an interactivity data structure and a tenured interactivity data structure are utilized in determining trends quickly.

Abstract: Systems, methods, and apparatuses for tracking trends at an online presence are described. In some embodiments, an interactivity data structure and a tenured interactivity data structure are utilized in determining trends quickly.

Abstract: In one embodiment, a set of user groups of a social-networking system may be accessed. A first subset of the user groups may be determined for a particular user of the social-networking system, based on one or more filtering criteria. A number of recommendation-source processes may be applied to the first subset to determine a number of second subsets of the first set. Each recommendation-source process may represent a particular recommendation source. The second subsets may be combined into a list of user groups. The list of user groups may be ranked, and sent to the particular user.

Abstract: Systems, methods, and apparatuses for tracking trends at an online presence are described. In some embodiments, an interactivity data structure and a tenured interactivity data structure are utilized in determining trends quickly.