Abstract: A method and related system for training a machine learning model using a federated learning structure by selectively sharing condensed data between devices includes operations to obtain datasets from client devices comprising a first client device and a second client device and updating a datasets subset to comprise a first dataset. The method further includes updating the datasets subset to comprise a second dataset based on a result indicating that a feature space distance between the first dataset and the second dataset satisfies a set of criteria and sending, to a third client device, the datasets subset comprising the first dataset and the second dataset. The method further includes obtaining, from the third client device, a set of model parameters that is derived from training based on the datasets subset and updating a server version of a machine learning model based on the set of model parameters.

Abstract: The present disclosure describes training machine learning models to determine one or more attributes and/or information associated with a location. The one or more attributes and/or information may be based on location traffic and/or connectivity information. After the machine learning model is trained and deployed, the machine learning model may determine one or more attributes and/or information about a location based on received location traffic and/or connectivity information. The machine learning model may receive a plurality of connectivity information associated with a location. Each of the plurality of connectivity information may correspond to a respective user device. Based on the connectivity information, the machine learning model may determine one or more attributes and/or information associated with the location. The one or more attributes and/or information may be provided to a user device in response to inquiries about the location.

Abstract: In some implementations, a machine learning system may receive a set of first data structures associated with a set of users, may receive a set of second data structures associated with a set of entities, and may receive event information associated with the set of users and the set of entities. The machine learning system may generate an embedding space that represents the set of first data structures, the set of second data structures, and the event information. The machine learning system may disregard a portion of the set of second data structures, using a condition, in order to generate a subset of second data structures. The machine learning system may identify at least one relevant entity, from the subset of second data structures, using the embedding space. The machine learning system may generate a communication, associated with the at least one relevant entity, and may output the communication.

Abstract: The present inventions is directed to systems and methods for automated stress testing of data models and comprises steps of generating, from an initial training dataset used for training a data model, a first data profile comprising a descriptive summary of the initial training dataset, generating a stress profile from an analysis of the data model and the initial training dataset used for training the data model, and modifying the first data profile by the stress profile to generate an altered data profile. A synthetic dataset may then be generated from the altered data profile. A stress performance output of the data model may then be used to identify weak points in the performance of the data model and improve the model's stress performance response using synthetic data.

Abstract: In some implementations, a machine learning system may generate a predicted event level, associated with an entity, based on event information associated with the entity. The machine learning system may determine that the predicted event level satisfies a threshold. The machine learning system may select, in response to determining that the predicted event level satisfies the threshold, a set of users. The machine learning system may transmit at least one communication to one or more user devices associated with the set of users.

Abstract: Systems and methods for detecting anomalous data updates in federated learning. In some aspects, the system retrieves a plurality of data updates and a corresponding plurality of data profiles during a first interval of time. The system generates a first data profile trend based on the plurality of data profiles, including an expectation value and a measure of variance. The system receives a subsequent data update and a corresponding subsequent data profile. The system determines a measure of deviation based on the expectation value of the first data profile trend and the subsequent data profile. The system generates an anomaly score based on the measure of deviation and the measure of variance. Based on the anomaly score, the system determines whether to label the subsequent data update as acceptable for inclusion in training data to generate a local model.

Abstract: Systems and methods for detecting anomalous data updates in federated learning. In some aspects, the system receives a plurality of data updates. Each data update contains data in a first real-valued space. The system selects a first function to project the plurality of data updates into a second real-valued space. The system selects a second function to partition the second real-valued space into a plurality of sectors. The system generates a plurality of sector datasets associated with the plurality of sectors. The system processes the plurality of sector datasets to generate a relational data structure. The system determines outliers in the relational data structure corresponding to anomalous data updates.

Abstract: In some implementations, an expert system may receive, from a user device, a data structure representing a project associated with a technical request. The expert system may map the data structure to a graph representation of a user space to determine a set of users. The expert system may determine a set of statuses, corresponding to the set of users, associated with availability of the set of users. The expert system may transmit, to the user device, instructions for a user interface including visual indicators for the set of users and the set of statuses. The expert system may receive, from the user device, an indication of a selected user from the set of users and may transmit a message addressed to the selected user.

Abstract: Methods and systems for scalable dataset content embedding for improved searchability. For example, the system may retrieve a first dataset from a first data source. The system may generate a first data profile of the first dataset. The system may generate a latent index of the first data profile based on processing the first data profile using a first embedding algorithm. The system may receive, via a user interface, a first request for a first text string. The system may generate an embedded request corresponding to the first request based on processing the first text string using the first embedding algorithm. The system may process the embedded request using the latent index. The system may generate for display, in the user interface, a result based on processing the embedded request using the latent index.

Abstract: Systems and methods for novel uses and/or improvements to data labeling applications, particularly data labeling applications involving sensitive data. As one example, systems and methods are described herein for preventing sensitive data leakage, using weak learner libraries, during label propagation.

Abstract: Systems and methods for deploying machine learning models trained on synthetic data generated based on a predicted future data drift are described herein. In some aspects, the system analyzes data profiles that include snapshots of data from different points in time from the same data stream to predict a future data drift. The system generates synthetic data based on the previous data profiles. The system determines, based on a first and second data profile, that a first data drift exceeds a threshold and updates a machine learning model. The system determines a predicted data drift and generates a synthetic snapshot of the data stream corresponding to a time in the future. The system generates a speculative machine learning model from updating the first updated machine learning model based on the synthetic snapshot. The system deploys the speculative machine learning model to replace the first updated machine learning model.

Abstract: A drive system for a vehicle that includes a transmission with three forward and three reverse gears. The transmission is optionally coupled to an electric motor as part of a fully electric or hybrid drive system. The transmission optionally includes a first and a second planetary gearset, a first and a second brake, and a clutch which may be separated into individual drive modules, optionally in separate detachable housings. A single electric motor may be provided with a single transmission, or multiple electric motors may be included to individually drive multiple separate tires, tracks, or other earth engaging elements. Single or multiple motors may be arranged upstream or downstream of the transmission, and they may be aligned parallel with or perpendicular to a drive axle.

Abstract: Systems and methods for federated learning including validation of training data on client devices before training models based on the training data. In some aspects, the system accesses local data to be used for training a local model on a client device. The system generates event metadata corresponding to events in the local data. The event metadata is based on validation data received from an external server. The validation data is inaccessible to a central server. The system determines to exclude a portion of the local data that is not validated by the event metadata and generates validated local data. The system trains the local model based on the validated local data. The system processes, using a data de-identification function, the event metadata to generate de-identified event metadata. The system transmits the local model and the de-identified event metadata to the central server.

Abstract: An apparatus for preparing a liquid preparation uses a two-part system. A first part includes a first reagent and a second part includes a second reagent. The first reagent and the second reagent react when mixed to form an active ingredient of the liquid preparation. The liquid preparation is a diluent. The apparatus includes a primary chamber for receiving a quantity of the first part, a quantity of the second part and a quantity of the diluent, a reagent inlet for admitting the first part and the second part to the primary chamber, a diluent inlet connectable to a diluent supply for admitting the diluent to the primary chamber, an outlet for releasing the liquid preparation from the primary chamber, and a flow controller to admit a pre-determined volume of diluent to the primary chamber in a pre-determined time period when the diluent inlet is connected to a diluent supply.

Abstract: An apparatus is disclosed for preparing a liquid preparation using a two-part system comprising a first part including a first reagent and a second part including a second reagent, wherein the first reagent and the second reagent react when mixed to form an active ingredient of the liquid preparation. The apparatus comprising a mixing funnel having a reaction chamber, an inlet region for admitting a quantity of the first part and a quantity of the second part into the reaction chamber, for example from a dispensing capsule, to form a reagent mixture in the reaction chamber in use, and an outlet comprising at least one restriction orifice for releasing the reagent mixture from the reaction chamber at a controlled rate, and a primary chamber arranged to receive the reagent mixture from the reaction chamber through the restriction orifice.

Abstract: An assembly (2) for dispensing liquids comprises: • a base member (8); • a cap (4) on the base member (8); and • a dispensing capsule (6) in the base member (8) and covered by the cap (4). The dispensing capsule (6) comprises: • a capsule body (7) having a first end (16) and a second end (18); • at least one wall (20) defining a first cavity (38) within the capsule body, the first cavity (38) having a first opening at the first end (16) of the capsule body; • at least one wall (22) defining a second cavity (40) within the capsule body, the second cavity (40) having a second opening at the first end (16) of the capsule body; • the first opening and the second opening provided with a seal member (64) which seals the first cavity (38) and the second cavity (40). The walls of the first cavity and the second cavity are collapsible if pressure is applied to the cavities from the second end (18) of the capsule body.

Abstract: An electric powertrain includes a first electric motor that has an uninterrupted connection with a drive shaft of a vehicle. The electric powertrain further includes a second electric motor that has an interruptible connection with the drive shaft. In one form, this interruptible connection includes a clutch. The electric powertrain further includes a first gear train in the form of a first planetary gear and a second gear train in the form of a second planetary gear. To provide a compact configuration, the first electric motor and second electric motor are arranged in a longitudinal orientation with the drive shaft.

Abstract: A vivarium measurement system for deployment in a vivarium environment and corresponding rooms that includes a microphone system, an at least one accelerometer in operational relationship to the microphone system, a data computational device that inputs data from the microphone system and the at least on accelerometer and wherein the data computational device comprises of a spectrum analyzer. The vivarium measurement system is optionally provided with an at least one alarm system that relays an alert when a threshold parameter is exceeded as determined by the spectrum analyzer.

Abstract: An electric powertrain includes a first electric motor that has an uninterrupted connection with a drive shaft of a vehicle. The electric powertrain further includes a second electric motor that has an interruptible connection with the drive shaft. In one form, this interruptible connection includes a clutch. The electric powertrain further includes a first gear train in the form of a first planetary gear and a second gear train in the form of a second planetary gear. To provide a compact configuration, the first electric motor and second electric motor are arranged in a longitudinal orientation with the drive shaft.