Abstract: Example implementations described herein involve, for data having incomplete labeling to generate a plurality of predictive maintenance models, processing the data through a multi-task learning (MTL) architecture including generic layers and task specific layers for the plurality of predictive maintenance models configured to conduct tasks to determine outcomes for one or more components associated with the data, each task specific layer corresponding to one of the plurality of predictive maintenance models; the generic layers configured to provide, to the task specific layers, associated data to construct each of the plurality of predictive maintenance models; and executing the predictive maintenance models on subsequently recorded data.

Abstract: A single unified machine learning model (e.g., a neural network) is trained to perform both supervised event predictions and unsupervised time-varying clustering for a sequence of events (e.g., a sequence representing a user behavior) using sequences of events for multiple users using a combined loss function. The unified model can then be used for, given a sequence of events as input, predict a next event to occur after the last event in the sequence and generate a clustering result by performing a clustering operation on the sequence of events. As part of predicting the next event, the unified model is trained to predict an event type for the next event and a time of occurrence for the next event. In certain embodiments, the unified model is a neural network comprising a recurrent neural network (RNN) such as an Long Short Term Memory (LSTM) network.

Abstract: A method can be implemented at one or more computing machines. The method can include receiving, using a server, time-series data corresponding to monitoring instrumentation in a medical care facility. The time-series data corresponds to a selected care recipient. The time-series data is stored in one or more data storage units. The time-series data includes data correlated with a plurality of regular time intervals. The method includes receiving, using a server, aperiodic data corresponding to clinical notes collected in the medical care facility and corresponding to the selected care recipient. The aperiodic data is stored in one or more data storage units. The aperiodic data includes a time stamp. The method includes generating, using a deep neural network and the time-series data and using a convolutional neural network (CNN) and the aperiodic data, a plurality of computer-generated data corresponding to management of the medical care facility or medical condition of the care recipient.

Abstract: Example implementations described herein involve, for data having incomplete labeling to generate a plurality of predictive maintenance models, processing the data through a multi-task learning (MTL) architecture including generic layers and task specific layers for the plurality of predictive maintenance models configured to conduct tasks to determine outcomes for one or more components associated with the data, each task specific layer corresponding to one of the plurality of predictive maintenance models; the generic layers configured to provide, to the task specific layers, associated data to construct each of the plurality of predictive maintenance models; and executing the predictive maintenance models on subsequently recorded data.

Abstract: A computing machine receives sensor data representing airflow or air pressure. The computing machine determines, using an artificial neural network, a current sleep stage corresponding to the sensor data. The current sleep stage is one of: wake, rapid eye movement (REM), light sleep, and deep sleep. The artificial neural network comprises a convolutional neural network (CNN), a recurrent neural network (RNN), and a conditional random field (CRF). The computing machine provides an output representing the current sleep stage.

Abstract: Techniques are provided for generating a logical application by grouping multiple physical distributions of an application for programming a plurality of electronic devices. A logical application can run separate commands through a single interface, lessening the number of connections needed between different user's electronic devices (e.g., smart phones, tablets, workstations, wearable computers) and a company's various servers. In certain embodiments, different physical distributions of the application may correspond to different operating systems versions of the application. Compiled code for each of these operating systems and device types is controlled and assigned to select users' devices from vastly different distribution architectures. In certain embodiments, a particular version (i.e.

Abstract: Techniques are provided for generating a logical application by grouping multiple physical distributions of an application for programming a plurality of electronic devices. A logical application can run separate commands through a single interface, lessening the number of connections needed between different user's electronic devices (e.g., smart phones, tablets, workstations, wearable computers) and a company's various servers. In certain embodiments, different physical distributions of the application may correspond to different operating systems versions of the application. Compiled code for each of these operating systems and device types is controlled and assigned to select users' devices from vastly different distribution architectures. In certain embodiments, a particular version (i.e.