Abstract: Techniques regarding determining hyperparameters for a differentially private federated learning process are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a hyperparameter advisor component that determines a hyperparameter for a model of a differentially private federated learning process based on a defined numeric relationship between a privacy budget, a learning rate schedule, and a batch size.

Abstract: Embodiments of the invention include a computer-implemented method that uses a processor system to access a first machine learning (ML) model. The first ML model has been trained using data of a first server. A first performance metric of the first ML model is determined using data of a second server. A benefit analysis is performed to determine a benefit of the first ML server and the second ML server participating in a federated learning system, where the benefit analysis includes using the first performance metric.

Abstract: According to one embodiment, a method, computer system, and computer program product for grouped federated learning is provided. The embodiment may include initializing a plurality of aggregation groups including a plurality of parties and a plurality of local aggregators. The embodiment may also include submitting a query to a first party from the plurality of parties. The embodiment may further include submitting an initial response to the query from the first party or a second party from the plurality of parties to a first local aggregator from the plurality of local aggregators. The embodiment may also include submitting a final response from the first local aggregator or a second local aggregator from the plurality of local aggregators to a global aggregator. The embodiment may further include building a machine learning model based on the final response.

Abstract: A method for tracking user interactions with an application includes: storing the application in a memory of a mobile device, the application being associated with an instrumented widget and a library, the widget including an event logger; executing the application and the widget; receiving, through a user interface of the mobile device, an input corresponding to the event logger of the widget; logging, by the library, the input corresponding to the event logger of the widget in the memory of the mobile device; filtering a plurality of events, including the input corresponding to the event logger of the widget, to manage what data is reported to a monitor; and transmitting the input corresponding to the event logger of the widget to a server as monitored data.

Abstract: A method of operating a mobile device includes displaying a user interface as an image, the user interface being composed of a plurality of widgets, storing a privacy policy identifying at least one of the widgets, capturing a screenshot image corresponding to the screenshot image, excluding the at least one of the widgets from the screenshot image to create a modified screenshot image, and transmitting the modified screenshot image over a network to a monitoring server.

Abstract: A method and system for virtualizing mobile device sensors includes requesting from a first mobile device a virtual connection with a mobile device having a specific type of sensor, configuring an operating system of the first mobile device to allow an application program to accept data by proxy from the available sensor of the second mobile device; receiving a response from a second mobile device having the sensor, establishing a trusted temporary communication connection between the first and second mobile devices, sending a control signal from an application program on the first mobile device to the second mobile for operating the sensor on the second mobile device and receiving device sensor data from the sensor on the second mobile device.

Abstract: Techniques regarding determining hyperparameters for a differentially private federated learning process are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise a hyperparameter advisor component that determines a hyperparameter for a model of a differentially private federated learning process based on a defined numeric relationship between a privacy budget, a learning rate schedule, and a batch size.

Abstract: Embodiments relate to training a machine learning model based on an iterative algorithm in a distributed, federated, private, and secure manner. Participating entities are registered in a collaborative relationship. The registered participating entities are arranged in a topology and a topological communication direction is established. Each registered participating entity receives a public additive homomorphic encryption (AHE) key and local machine learning model weights are encrypted with the received public key. The encrypted local machine learning model weights are selectively aggregated and distributed to one or more participating entities in the topology responsive to the topological communication direction. The aggregated sum of the encrypted local machine learning model weights is subjected to decryption with a corresponding private AHE key. The decrypted aggregated sum of the encrypted local machine learning model weights is shared with the registered participating entities.

Abstract: A machine-learning model generation method, system, and computer program product deciding, via a first algorithm, a machine-learning algorithm that is best for customer data, invoking the machine-learning algorithm to train a neural network model with the customer data, analyzing the neural network model produced by the training for an accuracy, and improving the accuracy by iteratively repeating the training of the neural network model until a customer-defined constraint is met, as determined by the first algorithm.

Abstract: A method and system for virtualizing mobile device sensors includes requesting from a first mobile device a virtual connection with a mobile device having a specific type of sensor, configuring an operating system of the first mobile device to allow an application program to accept data by proxy from the available sensor of the second mobile device; receiving a response from a second mobile device having the sensor, establishing a trusted temporary communication connection between the first and second mobile devices, sending a control signal from an application program on the first mobile device to the second mobile for operating the sensor on the second mobile device and receiving device sensor data from the sensor on the second mobile device.

Abstract: A method includes storing an application in a memory of a mobile device, wherein the application is associated with a widget including an event logger and a library managing the widget, executing, by the mobile device, the application and the widget, receiving, through a user interface of the mobile device, an input corresponding to the event logger of the widget, logging, by the library, the input corresponding to the event logger of the widget in the memory of the mobile device, and reporting, by the library, the input corresponding to the event logger of the widget to a remote monitoring server as monitored data.

Abstract: A method and system for virtualizing mobile device sensors includes requesting from a first mobile device a virtual connection with a mobile device having a specific type of sensor, receiving a response from a second mobile device having the sensor, establishing a trusted temporary communication connection between the first and second mobile devices, sending a control signal from an application program on the first mobile device to the second mobile for operating the sensor on the second mobile device and receiving device sensor data from the sensor on the second mobile device. The operating system of the first mobile device is coupled with the application program by a virtual machine monitor running on the first mobile device such that the first mobile device is a host machine and the second mobile device is a guest machine.

Abstract: A method for tracking user interactions with an application includes: storing the application in a memory of a mobile device, the application being associated with an instrumented widget and a library, the widget including an event logger; executing the application and the widget; receiving, through a user interface of the mobile device, an input corresponding to the event logger of the widget; logging, by the library, the input corresponding to the event logger of the widget in the memory of the mobile device; filtering a plurality of events, including the input corresponding to the event logger of the widget, to manage what data is reported to a monitor; and transmitting the input corresponding to the event logger of the widget to a server as monitored data.

Abstract: A method includes analyzing, on a first computing device, data from second computing device(s) of user interaction with a user interface of an application previously executed on the second computing device(s). The data corresponds to events caused by the user interaction with the user interface of the application. The first computing device generates representation(s) of the analyzed data and outputs the representation(s) of the user interaction. Another method includes capturing and logging, by a computing device, events caused by user interaction with a user interface of an application when the application is executed on the computing device. In response to a trigger, data comprising the captured and logged events is sent toward another computing device. Another method includes instrumenting a measurement library into an application to create an instrumented version of the application, and sending the instrumented application to computing device(s).

Abstract: A method includes analyzing, on a first computing device, data from second computing device(s) of user interaction with a user interface of an application previously executed on the second computing device(s). The data corresponds to events caused by the user interaction with the user interface of the application. The first computing device generates representation(s) of the analyzed data and outputs the representation(s) of the user interaction. Another method includes capturing and logging, by a computing device, events caused by user interaction with a user interface of an application when the application is executed on the computing device. In response to a trigger, data comprising the captured and logged events is sent toward another computing device. Another method includes instrumenting a measurement library into an application to create an instrumented version of the application, and sending the instrumented application to computing device(s).

Abstract: A method of collaboratively executing a task using first to N-th mobile devices in an ad-hoc network includes determining collaborative mobile devices out of the second to N-th mobile devices, receiving information corresponding to the collaborative mobile devices, dividing the task into first to M-th sub tasks, assigning each of the first to M-th sub tasks to at least one of the collaborative mobile devices, requesting executions of the first to M-th sub tasks to the collaborative mobile devices, and receiving execution results of the first to M-th sub tasks from the collaborative mobile devices. M and N are integers greater than one.

Abstract: A method and system for virtualizing mobile device sensors includes requesting from a first mobile device a virtual connection with a mobile device having a specific type of sensor, receiving a response from a second mobile device having the sensor, establishing a trusted temporary communication connection between the first and second mobile devices, sending a control signal from an application program on the first mobile device to the second mobile for operating the sensor on the second mobile device and receiving device sensor data from the sensor on the second mobile device. The operating system of the first mobile device is coupled with the application program by a virtual machine monitor running on the first mobile device such that the first mobile device is a host machine and the second mobile device is a guest machine.

Abstract: A method and system for virtualizing mobile device sensors includes requesting from a first mobile device a virtual connection with a mobile device having a specific type of sensor, receiving a response from a second mobile device having the sensor, establishing a trusted temporary communication connection between the first and second mobile devices, sending a control signal from an application program on the first mobile device to the second mobile for operating the sensor on the second mobile device and receiving device sensor data from the sensor on the second mobile device. The operating system of the first mobile device is coupled with the application program by a virtual machine monitor running on the first mobile device such that the first mobile device is a host machine and the second mobile device is a guest machine.

Abstract: A method includes storing an application in a memory of a mobile device, wherein the application is associated with a widget including an event logger and a library managing the widget, executing, by the mobile device, the application and the widget, receiving, through a user interface of the mobile device, an input corresponding to the event logger of the widget, logging, by the library, the input corresponding to the event logger of the widget in the memory of the mobile device, and reporting, by the library, the input corresponding to the event logger of the widget to a remote monitoring server as monitored data.

Abstract: An embodiment includes causing elements to be recognized that correspond to one or more browser-supported programming languages in an application. The recognized elements are those elements that can be converted to native user interface elements in the operating system and rendered on a display of a computing device. The elements in the one or more browser-supported programming languages are converted to native user interface elements. The native user interface elements are caused to be rendered on the display of the computing device. Methods, apparatus, software, and computer program products are disclosed.