Abstract: The disclosure herein describes elastically managing the execution of workers of multi-worker workloads on accelerator devices. A first worker of a workload is executed on an accelerator device during a first time interval. A first context switch point is identified when the first worker is in a first worker state. At the identified context switch point, a first memory state of the first worker is stored in a host memory and the accelerator device is configured to a second memory state of the second worker. The second worker is executed during a second time interval and a second context switch point is identified at the end of the second time interval when the second worker is in a state that is equivalent to the first worker state. During the intervals, collective communication operations between the workers are accumulated and, at the second context switch point, the accumulated operations are performed.

Abstract: The disclosure herein describes scheduling execution of artificial intelligence (AI) workloads in a cloud infrastructure platform. A global scheduler receives AI workloads associated with resource ticket values. The scheduler distributes the AI workloads to nodes based on balancing resource ticket values. Local schedulers of the nodes schedule AI workloads on resources based on the resource ticket values of the AI workloads. Based on scheduling the AI workloads, coordinator services of the local schedulers execute the distributed AI workloads on the infrastructure resources of the nodes. The disclosure further describes scheduling AI workloads based on priority tiers. A scheduler receives AI workloads, and each AI workload is associated with a priority tier indicative of a preemption priority while being executed. The AI workloads are scheduled for execution on a distributed set of nodes based on the priority tiers and then execute based on the scheduling.

Abstract: The disclosure herein describes platform-level checkpointing for deep learning (DL) jobs. The checkpointing is performed through capturing two kinds of state data: (i) GPU state (device state), and (ii) CPU state (host state). The GPU state includes GPU data (e.g., model parameters, optimizer state, etc.) that is located in the GPU and GPU context (e.g., the default stream in GPU, various handles created by the libraries such as DNN, Blas, etc.). Only a fraction of the GPU memory is copied because the checkpointing is done in a domain-aware manner. The “active” memory contains useful data like model parameters. To be able to capture the useful data, memory management is controlled to identify which parts of the memory are active. Also, to restore the destination GPU to the same context/state, a mechanism is used to capture such state-changing events on an original GPU and replayed on a destination GPU.

Abstract: A “Data Transfer Tool” extracts, labels and stores user data or information that may be confined within application silos during user interaction with arbitrary apps. The Data Transfer Tool enables sharing of this otherwise siloed data shared across and between authorized apps. The Data Transfer Tool provides a task-centric approach to define and extract structured and semantically meaningful information from source applications by providing multiple semantic models that are individually tailored to particular source apps. The Data Transfer Tool applies an accessibility API or the like of the OS in combination with the semantic model for the source app to scrape user entered or selected data or information from the source app. The Data Transfer Tool enables and creates new user experiences and increases user efficiency when interacting with various apps by making the scraped data or information available to subscribing destination apps approved to receive that data or information.

Abstract: This document relates to camera calibration. One example uses real-world distances and image coordinates of object features in images to determine multiple candidate camera calibrations for a camera.

Abstract: In some embodiments, an encryption system secures data using a homomorphic encryption. The encryption system encrypts a number by encrypting a number identifier of the number and combining the number and the encrypted number identifier using a mathematical operation to generate an encrypted number. The encrypted numbers may be stored at a server system along with their number identifiers. The server system can then generate an aggregation (e.g., sum) of the encrypted numbers and provide the aggregation, the encrypted numbers, and the number identifiers. The encryption system can then separate the aggregation of the numbers from the aggregation of the encrypted numbers using an inverse of the mathematical operation used in the encryption to effect removal of an aggregation of the encrypted number identifiers of the numbers from the aggregation of the encrypted numbers. The separated aggregation of the numbers is an aggregation of the plurality of the numbers.

Abstract: This document relates to camera calibration. One example uses real-world distances and image coordinates of object features in images to determine multiple candidate camera calibrations for a camera.

Abstract: A “Skip-Correlator” ensures carrier sensing symmetry between wireless devices of arbitrary transmission power levels, thereby enabling fair sharing of available spectrum of a wireless channel between the wireless devices. In various implementations, receivers of a wireless device receive wireless preambles from neighboring wireless transmitters. This preamble includes an indication of a transmission power level of the neighboring wireless transmitter. The wireless device then selects and correlates a subset of samples of the received preamble as a function of a transmission power level of that wireless device. The remaining samples of the preamble are skipped by the receiving wireless device, hence the use of the term “Skip-Correlator.” Further, the wireless device computes a total preamble energy of the correlated samples of the received preamble.

Abstract: A “Data Transfer Tool” extracts, labels and stores user data or information that may be confined within application silos during user interaction with arbitrary apps. The Data Transfer Tool enables sharing of this otherwise siloed data shared across and between authorized apps. The Data Transfer Tool provides a task-centric approach to define and extract structured and semantically meaningful information from source applications by providing multiple semantic models that are individually tailored to particular source apps. The Data Transfer Tool applies an accessibility API or the like of the OS in combination with the semantic model for the source app to scrape user entered or selected data or information from the source app. The Data Transfer Tool enables and creates new user experiences and increases user efficiency when interacting with various apps by making the scraped data or information available to subscribing destination apps approved to receive that data or information.

Abstract: An encryption system stores encrypted values for aggregation is provided. The encryption system accesses an input set with input values. For each distinct value in the input set of input values, the encryption system generates an output set with an encrypted output value corresponding to each input value. The encryption system sets the encrypted output value for a corresponding input value to an encryption of an indicator of a match when the corresponding input value is the same as that distinct value. Otherwise, the encryption sets the encrypted output value for the corresponding input value to an encryption of an indicator of no match. The encrypted output values can then be aggregated to generate an encrypted aggregation based on input values that match, and the encrypted aggregation can be decrypted to generated a decrypted aggregation based on the input values that match.

Abstract: A “Data Transfer Tool” extracts, labels and stores user data or information that may be confined within application silos during user interaction with arbitrary apps. The Data Transfer Tool enables sharing of this otherwise siloed data shared across and between authorized apps. The Data Transfer Tool provides a task-centric approach to define and extract structured and semantically meaningful information from source applications by providing multiple semantic models that are individually tailored to particular source apps. The Data Transfer Tool applies an accessibility API or the like of the OS in combination with the semantic model for the source app to scrape user entered or selected data or information from the source app. The Data Transfer Tool enables and creates new user experiences and increases user efficiency when interacting with various apps by making the scraped data or information available to subscribing destination apps approved to receive that data or information.

Abstract: An encryption system stores encrypted values for aggregation is provided. The encryption system accesses an input set with input values. For each distinct value in the input set of input values, the encryption system generates an output set with an encrypted output value corresponding to each input value. The encryption system sets the encrypted output value for a corresponding input value to an encryption of an indicator of a match when the corresponding input value is the same as that distinct value. Otherwise, the encryption sets the encrypted output value for the corresponding input value to an encryption of an indicator of no match. The encrypted output values can then be aggregated to generate an encrypted aggregation based on input values that match, and the encrypted aggregation can be decrypted to generated a decrypted aggregation based on the input values that match.

Abstract: A dynamic channel selection approach for wireless communication networks is provided by measuring an ineffective communication metric on a currently-used channel. The network can switch channels if the ineffective communication metric from a device on the network satisfies a channel selection condition. To change the wireless communication network to the new channel, all network devices in the wireless communication network are instructed to switch to the new channel. As network communications start on the new channel, the ineffective communication metric measurements begin again on the new channel. Communications on the new channel continue until the channel selection condition is satisfied on the new channel, at which point another random selection of a channel is executed. The process continues in such iterations during the operation of the network and its network devices.

Abstract: A “Skip-Correlator” ensures carrier sensing symmetry between wireless devices of arbitrary transmission power levels, thereby enabling fair sharing of available spectrum of a wireless channel between the wireless devices. In various implementations, receivers of a wireless device receive wireless preambles from neighboring wireless transmitters. This preamble includes an indication of a transmission power level of the neighboring wireless transmitter. The wireless device then selects and correlates a subset of samples of the received preamble as a function of a transmission power level of that wireless device. The remaining samples of the preamble are skipped by the receiving wireless device, hence the use of the term “Skip-Correlator.” Further, the wireless device computes a total preamble energy of the correlated samples of the received preamble.

Abstract: In some embodiments, an encryption system secures data using a homomorphic encryption. The encryption system encrypts a number by encrypting a number identifier of the number and combining the number and the encrypted number identifier using a mathematical operation to generate an encrypted number. The encrypted numbers may be stored at a server system along with their number identifiers. The server system can then generate an aggregation (e.g., sum) of the encrypted numbers and provide the aggregation, the encrypted numbers, and the number identifiers. The encryption system can then separate the aggregation of the numbers from the aggregation of the encrypted numbers using an inverse of the mathematical operation used in the encryption to effect removal of an aggregation of the encrypted number identifiers of the numbers from the aggregation of the encrypted numbers. The separated aggregation of the numbers is an aggregation of the plurality of the numbers.

Abstract: A method for managing data communication of a mobile device in a mobile network is provided. The method may include receiving a request, at a cloud proxy, to retrieve Internet data from the Internet. The request may originate from an application, and the Internet data may have a plurality unique objects. The method may also include aggregating the Internet data and two or more of the unique objects from the Internet. Furthermore, the method may include forwarding the Internet data and the two or more unique objects to the mobile device in one transmission.

Abstract: A multiple independent narrow-channel wireless network and method for transmitting and received data over a wireless network using a fragmented frequency spectrum. The system and method uses a plurality of narrow wireless channels obtained from splitting a wide wireless channel. Each narrow channel performs sending, receiving, and carrier sensing. Moreover, each narrow channel is independent such that data can be sent or received without any interference from other narrow channels and without synchronization. Embodiments of the system and method include a compound radio having a compound receiver and a compound transmitter. The compound transmitter includes an inter-radiolet symbol synchronization module, to permit use of a single inverse fast Fourier transform block, and a dynamically configurable filter array, to mitigate leakage between channels. The compound receiver uses fraction data rate processing to optimize efficiency.

Abstract: A dynamic channel selection approach for wireless communication networks is provided by measuring an ineffective communication metric on a currently-used channel. The network can switch channels if the ineffective communication metric from a device on the network satisfies a channel selection condition. To change the wireless communication network to the new channel, all network devices in the wireless communication network are instructed to switch to the new channel. As network communications start on the new channel, the ineffective communication metric measurements begin again on the new channel. Communications on the new channel continue until the channel selection condition is satisfied on the new channel, at which point another random selection of a channel is executed. The process continues in such iterations during the operation of the network and its network devices.

Abstract: A “Data Transfer Tool” extracts, labels and stores user data or information that may be confined within application silos during user interaction with arbitrary apps. The Data Transfer Tool enables sharing of this otherwise siloed data shared across and between authorized apps. The Data Transfer Tool provides a task-centric approach to define and extract structured and semantically meaningful information from source applications by providing multiple semantic models that are individually tailored to particular source apps. The Data Transfer Tool applies an accessibility API or the like of the OS in combination with the semantic model for the source app to scrape user entered or selected data or information from the source app. The Data Transfer Tool enables and creates new user experiences and increases user efficiency when interacting with various apps by making the scraped data or information available to subscribing destination apps approved to receive that data or information.

Abstract: One or more mobile devices and a cloud server computing device are utilized to provide Internet access to one or more client computing devices. One of the clients is designated as a Wi-Fi access point. The Wi-Fi access point implements a reverse-infrastructure Wi-Fi mode which tethers available mobile devices and other clients to the access point. The cloud server periodically computes an optimal number of the mobile devices to be used for data striping, and transmits a webpage to a requesting client by striping data associated with the webpage across the optimal number of the mobile devices.