Abstract: A system for predicting and summarizing medical events from electronic health records includes a computer memory storing aggregated electronic health records from a multitude of patients of diverse age, health conditions, and demographics including medications, laboratory values, diagnoses, vital signs, and medical notes. The aggregated electronic health records are converted into a single standardized data structure format and ordered arrangement per patient, e.g., into a chronological order. A computer (or computer system) executes one or more deep learning models trained on the aggregated health records to predict one or more future clinical events and summarize pertinent past medical events related to the predicted events on an input electronic health record of a patient having the standardized data structure format and ordered into a chronological order.

Abstract: The present disclosure is directed to an automated neural architecture search approach for designing new neural network architectures such as, for example, resource-constrained mobile CNN models. In particular, the present disclosure provides systems and methods to perform neural architecture search using a novel factorized hierarchical search space that permits layer diversity throughout the network, thereby striking the right balance between flexibility and search space size. The resulting neural architectures are able to be run relatively faster and using relatively fewer computing resources (e.g., less processing power, less memory usage, less power consumption, etc.), all while remaining competitive with or even exceeding the performance (e.g., accuracy) of current state-of-the-art mobile-optimized models.

Abstract: Generally, the present disclosure is directed to systems and methods that generate augmented training data for machine-learned models via application of one or more augmentation techniques to audiographic images that visually represent audio signals. In particular, the present disclosure provides a number of novel augmentation operations which can be performed directly upon the audiographic image (e.g., as opposed to the raw audio data) to generate augmented training data that results in improved model performance. As an example, the audiographic images can be or include one or more spectrograms or filter bank sequences.

Abstract: Generally, the present disclosure is directed to systems and methods that generate augmented training data for machine-learned models via application of one or more augmentation techniques to audiographic images that visually represent audio signals. In particular, the present disclosure provides a number of novel augmentation operations which can be performed directly upon the audiographic image (e.g., as opposed to the raw audio data) to generate augmented training data that results in improved model performance. As an example, the audiographic images can be or include one or more spectrograms or filter bank sequences.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: The present disclosure is directed to an automated neural architecture search approach for designing new neural network architectures such as, for example, resource-constrained mobile CNN models. In particular, the present disclosure provides systems and methods to perform neural architecture search using a novel factorized hierarchical search space that permits layer diversity throughout the network, thereby striking the right balance between flexibility and search space size. The resulting neural architectures are able to be run relatively faster and using relatively fewer computing resources (e.g., less processing power, less memory usage, less power consumption, etc.), all while remaining competitive with or even exceeding the performance (e.g., accuracy) of current state-of-the-art mobile-optimized models.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: An improved composition and method for stimulating plant growth and productivity, the composition of which can be adjusted to provide a preferred pH level that is ideal for growing plants in order to enhance the uptake of certain nutrients in growing such plants. The composition comprises combinations of Citric Acid, Disodium Hydrogen Phosphate or one of its hydtares, Boric Acid and Trisodium Phosphate or Trisodium Phosphate Dodecahydrate dissolved with a set quantity of water.

Abstract: As part of a dialog session between a user and an automated assistant, implementations can receive a stream of audio data that captures a spoken utterance including an assistant query, determine, based on processing the stream of audio data, a set of assistant outputs that are each predicted to be responsive to the assistant query, process, using large language model (LLM) output(s), the assistant outputs and context of the dialog session to generate a set of modified assistant outputs, and cause given modified assistant output, from among the set of modified assistant outputs, to be provided for presentation to the user in response to the spoken utterance. In some implementations, the LLM output(s) can be generated in an offline manner for subsequent use in an online manner. In additional or alternative implementations, the LLM output(s) can be generated in an online manner when the spoken utterance is received.

Abstract: The present disclosure is directed to an automated neural architecture search approach for designing new neural network architectures such as, for example, resource-constrained mobile CNN models. In particular, the present disclosure provides systems and methods to perform neural architecture search using a novel factorized hierarchical search space that permits layer diversity throughout the network, thereby striking the right balance between flexibility and search space size. The resulting neural architectures are able to be run relatively faster and using relatively fewer computing resources (e.g., less processing power, less memory usage, less power consumption, etc.), all while remaining competitive with or even exceeding the performance (e.g., accuracy) of current state-of-the-art mobile-optimized models.

Abstract: A system includes a plurality of secure gateways that each use a plurality of datasets to determine how to process messages between devices on a network and websites on the internet. A version control server in the system automatically sends a dataset to each secure gateway in the plurality of secure gateways.

Abstract: In an embodiment, a system for synchronizing the rotation of multiple mainframes of an airship includes multiple belt drive systems configured to mechanically rotate the mainframes, a central control system for sending a timing instruction to cause the mainframes to rotate synchronously about their respective rotational axis, wherein the mainframes are axis-aligned about their respective rotational axes and the timing instruction specifies a desired angular displacement of the mainframes, and multiple control units for controlling the belt drive systems to rotate the mainframes, respectively, wherein, for each mainframe, the associated control unit is configured to: receive the timing instruction from the central control system; determine, according to the timing instruction, a rotation instruction based on a size of the mainframe and the desired angular displacement; and instruct the belt drive system controlled by the control unit to rotate the mainframe based on the rotation instruction.

Abstract: A system for predicting and summarizing medical events from electronic health records includes a computer memory storing aggregated electronic health records from a multitude of patients of diverse age, health conditions, and demographics including medications, laboratory values, diagnoses, vital signs, and medical notes. The aggregated electronic health records are converted into a single standardized data structure format and ordered arrangement per patient, e.g., into a chronological order. A computer (or computer system) executes one or more deep learning models trained on the aggregated health records to predict one or more future clinical events and summarize pertinent past medical events related to the predicted events on an input electronic health record of a patient having the standardized data structure format and ordered into a chronological order.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: In an optical filter for sunglasses, having a transmittance of less than 20% for light wavelengths from 400 nm to 650 nm, the transmission spectrum comprises a local transmission maximum in each of the light wavelength ranges between 440 nm and 470 nm and also between 570 nm and 590 nm and a local transmission maximum in the light wavelength range between 600 nm and 620 nm, wherein the transmittance below the connecting line of the local transmission maximum between 440 nm and 470 nm and the local transmission maximum between 570 nm and 590 nm has an essentially convex overall curve with a breadth of variation less than 8%.

Abstract: In an embodiment, a system for synchronizing the rotation of multiple mainframes of an airship includes multiple belt drive systems configured to mechanically rotate the mainframes, a central control system for sending a timing instruction to cause the mainframes to rotate synchronously about their respective rotational axis, wherein the mainframes are axis-aligned about their respective rotational axes and the timing instruction specifies a desired angular displacement of the mainframes, and multiple control units for controlling the belt drive systems to rotate the mainframes, respectively, wherein, for each mainframe, the associated control unit is configured to: receive the timing instruction from the central control system; determine, according to the timing instruction, a rotation instruction based on a size of the mainframe and the desired angular displacement; and instruct the belt drive system controlled by the control unit to rotate the mainframe based on the rotation instruction.

Abstract: Example aspects of the present disclosure are directed to systems and methods for learning data augmentation strategies for improved object detection model performance. In particular, example aspects of the present disclosure are directed to iterative reinforcement learning approaches in which, at each of a plurality of iterations, a controller model selects a series of one or more augmentation operations to be applied to training images to generate augmented images. For example, the controller model can select the augmentation operations from a defined search space of available operations which can, for example, include operations that augment the training image without modification of the locations of a target object and corresponding bounding shape within the image and/or operations that do modify the locations of the target object and bounding shape within the training image.

Abstract: Generally, the present disclosure is directed to systems and methods that generate augmented training data for machine-learned models via application of one or more augmentation techniques to audiographic images that visually represent audio signals. In particular, the present disclosure provides a number of novel augmentation operations which can be performed directly upon the audiographic image (e.g., as opposed to the raw audio data) to generate augmented training data that results in improved model performance. As an example, the audiographic images can be or include one or more spectrograms or filter bank sequences.

Abstract: An uninterruptible power supply for providing an output power signal to a load comprises a ferroresonant transformer, a resonant capacitor, and an inverter. The resonant capacitor is operatively connected to the ferroresonant transformer. The inverter is operatively connected to the ferroresonant transformer. The inverter is configured to generate the output power signal based on at least one inverter control signal such that the output power signal is a quasi square wave having at least one change of phase and an upper limit. The at least one inverter control signal is held in an OFF state during at least a portion of the at least one change of phase, pulse-width modulated during at least a portion of the at least one change of phase, and held in an ON state when the output power signal is at the upper limit.

Abstract: A ferroresonant transformer system has a core, a shunt, first and second input windings, an inverter winding, a tank winding, a resonant capacitor, an output capacitor, and a plurality of switches. The tank winding defines a plurality of switch tap locations and at least two output tap locations. The resonant capacitor is connected across at least a portion of the tank winding. Each switch is operatively connectable between one of the switch tap locations and the resonant capacitor. Each output terminal is operatively connected to one of the at least two output tap locations.