Abstract: Systems, methods, and computer-readable media are disclosed for parallel stochastic gradient descent using linear and non-linear activation functions. One method includes: receiving a set of input examples; receiving a global model; and learning a new global model based on the global model and the set of input examples by iteratively performing the following steps: computing a plurality of local models having a plurality of model parameters based on the global model and at least a portion of the set of input examples; computing, for each local model, a corresponding model combiner based on the global model and at least a portion of the set of input examples; and combining the plurality of local models into the new global model based on the current global model and the plurality of corresponding model combiners.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed to optimizing the generation, evaluation, and selection of tensor circuit specifications for a tensor circuit to perform homomorphic encryption operations on encrypted data. A computing device having an improved compiler and runtime configuration can obtain a tensor circuit and associated schema. The computing device can map the obtained tensor circuit to an equivalent tensor circuit, adapted to perform fully homomorphic encryption (FHE) operations, and instantiated based on the obtained associated scheme. The computing device can then monitor a flow of data through the equivalent FHE-adapted tensor circuit utilizing various tensor circuit specifications determined therefor.

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. The symbolic representations can be used to combine the local models. The global model can determine a likelihood, given a new data instance of a feature set, that a user performs a computer interaction with the content element. For instance, the system can use the model to provide search results in response to a search query submitted by a user. Or, the system can use the model to make a recommendation or suggestion to a user in response to a request for content (e.g., display a targeted advertisement, suggest a news story, etc.).

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood of a course of action being successful for an organization. For example, the course of action can be a purchase of a security or a business operation strategy. In another example, the course of action can be a type of medical treatment for a patient.

Abstract: Described herein is a system transmits and combines local models, that individually include a set of local parameters computed via stochastic gradient descent (SGD), into a global model that includes a set of global model parameters. The local models are computed in parallel at different geographic locations (e.g., different instances of computing infrastructure) along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood that at least a portion of current and/or recently received data traffic is illegitimate data traffic that is associated with a cyber attack. In some instances, the system can implement a remedial action to mitigate the effects of the cyber attack on computing infrastructure.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed to optimizing the generation, evaluation, and selection of tensor circuit specifications for a tensor circuit to perform homomorphic encryption operations on encrypted data. A computing device having an improved compiler and runtime configuration can obtain a tensor circuit and associated schema. The computing device can map the obtained tensor circuit to an equivalent tensor circuit, adapted to perform fully homomorphic encryption (FHE) operations, and instantiated based on the obtained associated scheme. The computing device can then monitor a flow of data through the equivalent FHE-adapted tensor circuit utilizing various tensor circuit specifications determined therefor.

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood of a monitored resource or a user of the monitored resource experiencing a problem with respect to performance or completion of one or more operations. The system can also implement an action to assist in resolving or avoiding the problem.

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood of a course of action being successful for an organization. For example, the course of action can be a purchase of a security or a business operation strategy. In another example, the course of action can be a type of medical treatment for a patient.

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. The symbolic representations can be used to combine the local models. The global model can determine a likelihood, given a new data instance of a feature set, that a user performs a computer interaction with the content element. For instance, the system can use the model to provide search results in response to a search query submitted by a user. Or, the system can use the model to make a recommendation or suggestion to a user in response to a request for content (e.g., display a targeted advertisement, suggest a news story, etc.).

Abstract: Described herein is a system transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations (e.g., different instances of computing infrastructure) along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood that at least a portion of current and/or recently received data traffic is illegitimate data traffic that is associated with a cyber attack. In some instances, the system can implement a remedial action to mitigate the effects of the cyber attack on computing infrastructure.

Abstract: Described herein is a system that transmits and combines local models, that individually comprise a set of local parameters computed via stochastic gradient descent (SGD), into a global model that comprises a set of global model parameters. The local models are computed in parallel at different geographic locations along with symbolic representations. Network transmission of the local models and the symbolic representations, rather than transmission of the large training data subsets processed to compute the local models and symbolic representations, conserves resources and decreases latency. The global model can then be used as a model to determine a likelihood of a monitored resource or a user of the monitored resource experiencing a problem with respect to performance or completion of one or more operations. The system can also implement an action to assist in resolving or avoiding the problem.

Abstract: Systems, methods, and computer-readable media are disclosed for parallel stochastic gradient descent using linear and non-linear activation functions. One method includes: receiving a set of input examples; receiving a global model; and learning a new global model based on the global model and the set of input examples by iteratively performing the following steps: computing a plurality of local models having a plurality of model parameters based on the global model and at least a portion of the set of input examples; computing, for each local model, a corresponding model combiner based on the global model and at least a portion of the set of input examples; and combining the plurality of local models into the new global model based on the current global model and the plurality of corresponding model combiners.

Abstract: The described implementations relate to analysis of computing programs. One implementation provides a technique that can include accessing values of input variables that are processed by test code and runtime values that are produced by the test code while processing the input variables. The technique can also include modeling relationships between the runtime values and the values of the input variables. The relationships can reflect discontinuous functions of the input variables.

Abstract: Described is a probabilistic concurrency testing mechanism for testing a concurrent software program that provides a probabilistic guarantee of finding any concurrent software bug at or below a bug depth (that corresponds to a complexity level for finding the bug). A scheduler/algorithm inserts priority lowering points into the code and runs the highest priority thread based upon initially randomly distributed priorities. When that thread reaches a priority lowering point, its priority is lowered to a value associated (e.g., by random distribution) with that priority lowering point, whereby a different thread now has the currently highest priority. That thread is run until its priority is similarly lowered, and so on, whereby all schedules needed to find a concurrency bug are run.

Abstract: The subject disclosure relates to effective dynamic monitoring of an application executing in a computing system by increasing concurrency coverage. A set of dynamic checks are linked to an application by mechanisms that enable the dynamic checks to monitor behavior of the application at runtime. As additionally described herein, concurrency fuzzing is applied to the application to randomize thread schedules of the application, thus increasing a number of disparate concurrency scenarios of the application observed by the plurality of dynamic checks.

Abstract: A checking system is described for determining whether a component is thread safe in the course of interacting with two or threads in a client environment. The checking system uses a manual, automatic, or semi-automatic technique to generate a test. The checking system then defines a set of coarse-grained observations for the test, in which the component is assumed to exhibit linearizability when interacting with threads. The set of coarse-grained observations may include both complete and “stuck” histories. The checking system then generates a set of fine-grained observations for the tests; here, the checking system makes no assumptions as to the linearizability of the component. The checking system identifies potential linearizability errors as those entries in the set of fine-grained observations that have no counterpart entries in the set of coarse-grained observations. The checking system may rely on a stateless model checking module to perform its functions.

Abstract: The described implementations relate to analysis of computing programs. One implementation provides a technique that can include accessing values of input variables that are processed by test code and runtime values that are produced by the test code while processing the input variables. The technique can also include modeling relationships between the runtime values and the values of the input variables. The relationships can reflect discontinuous functions of the input variables.

Abstract: The subject disclosure relates to effective dynamic monitoring of an application executing in a computing system by increasing concurrency coverage. A set of dynamic checks are linked to an application by mechanisms that enable the dynamic checks to monitor behavior of the application at runtime. As additionally described herein, concurrency fuzzing is applied to the application to randomize thread schedules of the application, thus increasing a number of disparate concurrency scenarios of the application observed by the plurality of dynamic checks.

Abstract: A checking system is described for determining whether a component is thread safe in the course of interacting with two or threads in a client environment. The checking system uses a manual, automatic, or semi-automatic technique to generate a test. The checking system then defines a set of coarse-grained observations for the test, in which the component is assumed to exhibit linearizability when interacting with threads. The set of coarse-grained observations may include both complete and “stuck” histories. The checking system then generates a set of fine-grained observations for the tests; here, the checking system makes no assumptions as to the linearizability of the component. The checking system identifies potential linearizability errors as those entries in the set of fine-grained observations that have no counterpart entries in the set of coarse-grained observations. The checking system may rely on a stateless model checking module to perform its functions.

Abstract: Described is a probabilistic concurrency testing mechanism for testing a concurrent software program that provides a probabilistic guarantee of finding any concurrent software bug at or below a bug depth (that corresponds to a complexity level for finding the bug). A scheduler/algorithm inserts priority lowering points into the code and runs the highest priority thread based upon initially randomly distributed priorities. When that thread reaches a priority lowering point, its priority is lowered to a value associated (e.g., by random distribution) with that priority lowering point, whereby a different thread now has the currently highest priority. That thread is run until its priority is similarly lowered, and so on, whereby all schedules needed to find a concurrency bug are run.