Abstract: In general, techniques are described by which to perform memory efficient patching for computing devices. A server computing device comprising a memory and a processor may be configured to perform the techniques. The memory may store first assets that form an unpatched application and second assets that form a patched application. The processor may virtualize the first assets to obtain a single first virtual asset and obtain a single second virtual asset that represents the second assets. The processor may obtain, based on differences between the single first virtual asset and the single second virtual asset, a patch identifying how to update the single first virtual asset to obtain the single second virtual asset. The processor may next segment the patch into a plurality of segments and output to a user computing device a single segment from the plurality of segments for individual application by the user computing device.

Abstract: Systems and methods are described that use software diversification techniques to improve the security of mobile applications. Embodiments of the disclosed systems and methods may, among other things, facilitate secure application distribution through deployment of diverse of applications in an application distribution channel. Software diversification consistent with certain disclosed embodiments may mitigate large-scale automated circumvention of security protections by presenting attacking malware moving and/or otherwise unpredictable diverse targets.

Abstract: The subject technology provides receiving a neural network (NN) model to be executed on a target platform, the NN model including multiple layers that include operations and some of the operations being executable on multiple processors of the target platform. The subject technology further sorts the operations from the multiple layers in a particular order based at least in part on grouping the operations that are executable by a particular processor of the multiple processors. The subject technology determines, based at least in part on a cost of transferring the operations between the multiple processors, an assignment of one of the multiple processors for each of the sorted operations of each of the layers in a manner that minimizes a total cost of executing the operations. Further, for each layer of the NN model, the subject technology includes an annotation to indicate the processor assigned for each of the operations.

Abstract: A system and method for detecting code initialization actions is utilized for performing static analysis of an application code including an external initialization component. The method includes receiving an application code including an external initialization code component, the application code deployed in a cloud computing environment; detecting in the application code an anchor point, the anchor point including a call to another code object; inserting a hook into the application code based on a location of the anchor point in the application code; and receiving a result from the hook in response to executing the application code.

Abstract: A data exchange system comprises an electronic smoking device that includes identifying data and a database configured to i) receive the identifying data from the electronic smoking device, ii) receive user data for a user of the electronic smoking device from a processor, iii) associate the identifying data with user data to generate a user record, and iv) store the user record.

Abstract: Methods, systems, and computer-readable media for debugging and profiling of machine learning model training are disclosed. A machine learning analysis system receives data associated with training of a machine learning model. The data was collected by a machine learning training cluster. The machine learning analysis system performs analysis of the data associated with the training of the machine learning model. The machine learning analysis system detects one or more conditions associated with the training of the machine learning model based at least in part on the analysis. The machine learning analysis system generates one or more alarms describing the one or more conditions associated with the training of the machine learning model.

Abstract: A computer that dynamically maps traffic to different versions of software is described. During operation, the computer may receive first traffic from a first tenant and second traffic from a second tenant, where the first traffic and the second traffic are addressed to a common computing environment that supports multiple, different tenants. Then, the computer may identify the first tenant based at least in part on first information included in the first traffic, and may identify the second tenant based at least in part on second information included in the second traffic. Moreover, the computer may dynamically map the first traffic to a first version of software and the second traffic to a second version of the software, where the dynamic mapping occurs at runtime. Note that the computer may concurrently execute the first version of the software and the second version of the software.

Abstract: A testing system, tool and method for determining anonymization process quality are disclosed. The testing tool includes a processor, a communications module coupled to the processor, and a memory coupled to the processor. The memory stores computer executable instructions that when executed by the processor cause the processor to obtain a first dataset comprising data to be anonymized. The processor obtains a second dataset comprising data resulting from the first dataset being subjected to an anonymization process by a testing platform to anonymize a portion of the first dataset and generates a visual representation of the first dataset. The processor generates a second visual representation of the second dataset, the second visual representation reflecting at least in part a transformation from the first dataset to the second dataset. The processor generates an output including the first and second visual representations, the output being illustrative of the transformation.

Abstract: According to examples, an apparatus may include a processor that may delete portions of firmware instructions responsive to an indication that the portions should be disabled. To facilitate the foregoing, the firmware instructions may be deployed in a segmented architecture stored in respective regions of a storage device. The regions may include a metadata region, a main region, and excludable regions. The metadata region may store metadata that describes the structure of the firmware instructions and/or the various other regions. The main region may store core firmware instructions that may not be deleted. Each excludable region may store respective excludable firmware instructions. Each excludable firmware instructions may be associated with a flag that indicates whether or not the instructions should be disabled. If so, the corresponding excludable region in the storage device is identified and the contents may be removed, permanently disabling the excludable firmware instructions that were stored there.

Abstract: Disclosed herein are a Docker image creation apparatus and method. The Docker image creation apparatus may include one or more processors, and execution memory for storing at least one program that is executed by the one or more processors, wherein the at least one program is configured to create a Docker image based on a previous Docker file, and executing a push command to store the created the Docker file to a registry, wherein the Docker file is created from a partial Docker file of the previous Docker file, wherein the Docker image is created from a verified partial Docker file, after the partial Docker file is verified.

Abstract: A device configured to identify automated process instructions for a first network device. The automated process instructions include instructions for performing a series of actions on the first network device. The device is further configured to identify a first time instance for executing the automated process instructions that is before a time interval associated with a system update for the first network device, to identify a second time instance corresponding with a start time of the time interval associated with the system update for the first network device, and to identify a third time instance corresponding with an end time of the time interval associated with the system update for the first network device.

Abstract: A device of updating library required by a testing program for testing and a method thereof are disclosed. A system configuration file of an old testing device is modified to make the old testing device enter a testing mode when booting, so that library supporting the testing program of an application is updated in the testing mode; when the application is executed, the testing program is connected to the to-be-tested device for performing testing, so that the old testing device can be updated through network without dedicated hardware, and the technical effect of reducing the time and labor cost for updating the old testing device can be achieved.

Abstract: The present disclosure concerns methods and systems for registering dynamically created packaged applications with an operating system. A dynamically created packaged application may not include a way to authenticate the packaged application but may declare a dependency on a host package. The host package may include a verified signature and be registered with the operating system. The host package may include a host runtime that references an executable included in the host package. The packaged application may not include executable files but may reference the host runtime. The operating system may register the packaged application based in part on its dependency on the host package. When the packaged application is activated, the packaged application may invoke the host runtime and the executable but do so with an application identity of the packaged application. The application identity may be different from a host runtime identity of the host runtime.

Abstract: Techniques for feature-based deployment pipelines are disclosed. The feature-based deployment pipelines use a service file as well as other manifest files to set up a repository, a continuous integration service to initiate testing, and a container orchestration system to build images and establish version control over the system. The feature-based deployment pipelines invoke sequential stages to enable feature branches of an application to be fully tested before proceeding to a succeeding stage. At each stage, relevant stakeholders are able to evaluate the new features before they become part of the master image of the tested application. A variety of validation and performance tests are conducted at each stage, resulting in a fully vetted application available for a consumer.

Abstract: Systems and methods for scheduling and deploying patches to electronic device assets are provided. A patch management system identifies scheduling for patching and causes a patch scripting service to apply patching tags to the electronic device assets. A patch deployment service deploys patches based upon the applied tags.

Abstract: Techniques are described herein for translating a source code snippet from a first programming language to a second programming language independently of sequence-to-sequence decoding. In various implementations, the source code snippet written in the first programming language may be processed using an encoder portion of a transformer network to generate an embedding of the source code snippet. The embedding of the source code snippet may be processed using an all-pair attention layer to generate an attended embedding of the source code snippet. The attended embedding of the source code snippet may be processed using an output layer to generate, by way of a single transformation of the attended embedding of the source code snippet, data indicative of a translation of the source code snippet in the second programming language.

Abstract: Technology is disclosed for establishing and administering multiple virtual machines, each with an audio, video and control (AVC) operating system (OS). The technology can also establish and administer cloud based AVC OSs. A server implementing this technology can perform real-time AVC processing, alongside soft and non-real-time processing and can host multiple, independent, virtual AVC OSs. Each AVC OS can perform the processing for an AVC setup. Each of the AVC OSs can be operated by a corresponding virtual machine controlled by a hypervisor running on the server. A cloud based AVC OS can perform processing for a corresponding remote AVC setup comprising multiple AVC devices. An AVC routing system can cause AVC signals from a particular AVC setup to reach a corresponding cloud AVC OS and conversely can cause signals from an AVC OS to reach the correct destination device.

Abstract: A debug information abstraction unit (32) of a proxy server (30) adds, when a predetermined rule of an extension program is embodied into a program, a program for outputting debug information regarding the predetermined rule to the embodied program. Then, a debug information transmission unit (41b) of a terminal device (40) transmits, to a debug support device (10), the debug information regarding the predetermined rule for which the program has been added.

Abstract: Disclosed herein are methods, systems, and computer-readable media for generating natural language based on computer code input. In an embodiment, a method may comprise one or more of: accessing a docstring generation model configured to generate docstrings from computer code; receiving one or more computer code samples; generating, using the docstring generation model and based on the received one or more computer code samples, one or more candidate docstrings representing natural language text, each of the one or more candidate docstrings being associated with at least a portion of the one or more computer code samples; identifying at least one of the one or more candidate docstrings that provides an intent of the at least a portion of the one or more computer code samples; and/or outputting, via a user interface, the at least one identified docstring with the at least a portion of the one or more computer code samples.

Abstract: Systems and methods for automated code analysis and tagging are disclosed. In one embodiment, a method for automated code analysis and tagging may include: (1) receiving, by a code annotation computer program executed by a computer processor, a training code snippet from a training codebase; (2) parsing, by the code annotation computer program, the training code snippet into a data structure; (3) quantifying, by the code annotation computer program, the data structure, (4) parsing, by the code annotation computer program, a docstring associated with the training code snippet into a plurality of keywords; (5) quantifying, by the code annotation computer program, the plurality of keywords; and (6) training, by the code annotation computer program, a code annotation model based on a similarity between the quantified data structure and a smoothing parameter for a Dirichlet prior smoothing estimate.