Abstract: This disclosure describes systems and techniques for detecting events, determining a result of each respective event using a first hypothesis source, and calculating a likelihood that a second (and/or additional) hypothesis source would determine the same result of the respective event. The calculated likelihood may then be used to be determine whether to request that the second hypothesis source determine the result of the event, determine an amount of resources of the second hypothesis source to use to make this determination, and/or like.

Abstract: This disclosure describes systems and techniques for detecting events, determining a result of each respective event using a first hypothesis source, and calculating a likelihood that a second (and/or additional) hypothesis source would determine the same result of the respective event. The calculated likelihood may then be used to be determine whether to request that the second hypothesis source determine the result of the event, determine an amount of resources of the second hypothesis source to use to make this determination, and/or like.

Abstract: A distributed system for training a classifier is provided. The system comprises machine learning (ML) workers and a parameter server (PS). The PS is configured for parallel processing to provide the model to each of the ML workers, receive model updates from each of the ML workers, and iteratively update the model using each model update. The PS contains gradient datasets associated with a respective ML worker, for storing a model-update-identification (delta-M-ID) indicative of the computed model update and the respective model update, a global dataset that stores, the delta-M-ID, an identification of the ML worker (ML-worker-ID) that computed the model update, and a model version that marks a new model in PS that is computed from merging the model update with a previous model in PS; and a model download dataset that stores the ML-worker-ID and the model version of each transmitted model.

Abstract: A computer system for distributed training of a machine learning model comprising a BSP system, at least one machine learning module, and a shared memory module. The BSP system includes a central BSP control module and at least one local BSP module. The central BSP control module is configured to instruct the at least one local BSP module to store, in its associated shared memory module, a local model. The at least one machine learning module is configured to read, from its associated shared memory module, the local model, compute a gradient based on the local model, and aggregate the gradient immediately after its computation into an aggregated gradient in its associated shared memory module. The central BSP control module is further configured to instruct the at least one local BSP module to periodically read out its associated shared memory module.

Abstract: A distributed system for training a classifier is provided. The system comprises machine learning (ML) workers and a parameter server (PS). The PS is configured for parallel processing to provide the model to each of the ML workers, receive model updates from each of the ML workers, and iteratively update the model using each model update. The PS contains gradient datasets associated with a respective ML worker, for storing a model-update-identification (delta-M-ID) indicative of the computed model update and the respective model update, a global dataset that stores, the delta-M-ID, an identification of the ML worker (ML-worker-ID) that computed the model update, and a model version that marks a new model in PS that is computed from merging the model update with a previous model in PS; and a model download dataset that stores the ML-worker-ID and the model version of each transmitted model.

Abstract: A system determines user context (UC) keywords associated with a context of a user of a computing device based on extracting words from context items associated with the user. The system also determines search provider (SP) keywords for each of a plurality of search providers, the SP keywords associated with a respective textual content (e.g., documents) of each of the plurality of search providers. Determining the SP keywords for a search provider may include calculating a term frequency-inverse document frequency (tf-idf) score for each word of each content item of the textual content of the search provider. The system then selects a search provider (or several) from the plurality of search providers based on a number of UC keywords that match the search provider's SP keywords being greater than a threshold number. The system then generates a query for the selected search provider based on the matching UC keywords.

Abstract: According to some embodiments, a suggestion platform associated with a content aggregation system may access a plurality of data context trigger criteria associated with potential business information content elements. Each data context trigger criteria may include, for example: (i) a data locator associated with a business information data structure, (ii) an operator, and (iii) a threshold. For each data context trigger criteria, it may be automatically determined if a value in the business information data structure satisfies the data context trigger criteria based on the operator and the threshold. When a data context trigger criteria is satisfied, it may be automatically suggested to a user that the potential business information content element associated with that data context trigger criteria be added to a user display of the content aggregation system.

Abstract: According to some embodiments, a suggestion platform associated with a content aggregation system may access a plurality of data context trigger criteria associated with potential business information content elements. Each data context trigger criteria may include, for example: (i) a data locator associated with a business information data structure, (ii) an operator, and (iii) a threshold. For each data context trigger criteria, it may be automatically determined if a value in the business information data structure satisfies the data context trigger criteria based on the operator and the threshold. When a data context trigger criteria is satisfied, it may be automatically suggested to a user that the potential business information content element associated with that data context trigger criteria be added to a user display of the content aggregation system.

Abstract: According to some embodiments, a suggestion platform associated with a content aggregation system may access a plurality of data context trigger criteria associated with potential business information content elements. Each data context trigger criteria may include, for example: (i) a data locator associated with a business information data structure, (ii) an operator, and (iii) a threshold. For each data context trigger criteria, it may be automatically determined if a value in the business information data structure satisfies the data context trigger criteria based on the operator and the threshold. When a data context trigger criteria is satisfied, it may be automatically suggested to a user that the potential business information content element associated with that data context trigger criteria be added to a user display of the content aggregation system.

Abstract: Methods, systems, and computer-readable storage media for configuring a request-centric system. Implementations include actions of monitoring operation of the request-centric system to determine a workload of the request-centric system, wherein the workload includes a plurality of requests, determining a mapping function based on a utility function, the workload and a configuration, the mapping function mapping the workload and the configuration into the utility function, determining an optimal configuration based on the mapping function, the optimal configuration allocating resources of the request-centric system to achieve one or more predefined goals, and applying the optimal configuration to the request-centric system.

Abstract: In one embodiment, a method includes: receiving data representing a plurality of corpora, each of the plurality of corpora including a set of documents; receiving data representing terms that appear in the corpora; for each one of the terms, determining a plurality of inverse document frequency values each associated with a respective one of the plurality of corpora; receiving data representing a subset of the terms that also appear in a document; for each term in the subset, determining a term frequency for the term in the document; and for each term in the subset, determining, an augmented term frequency-inverse document frequency value based on: (i) the term frequency, and (ii) the plurality of inverse document frequency values that were determined for the term in the subset.

Abstract: According to some embodiments, a suggestion platform associated with a content aggregation system may access a plurality of data context trigger criteria associated with potential business information content elements. Each data context trigger criteria may include, for example: (i) a data locator associated with a business information data structure, (ii) an operator, and (iii) a threshold. For each data context trigger criteria, it may be automatically determined if a value in the business information data structure satisfies the data context trigger criteria based on the operator and the threshold. When a data context trigger criteria is satisfied, it may be automatically suggested to a user that the potential business information content element associated with that data context trigger criteria be added to a user display of the content aggregation system.

Abstract: In one embodiment, a method includes: receiving data representing a plurality of corpora, each of the plurality of corpora including a set of documents; receiving data representing terms that appear in the corpora; for each one of the terms, determining a plurality of inverse document frequency values each associated with a respective one of the plurality of corpora; receiving data representing a subset of the terms that also appear in a document; for each term in the subset, determining a term frequency for the term in the document; and for each term in the subset, determining, an augmented term frequency-inverse document frequency value based on: (i) the term frequency, and (ii) the plurality of inverse document frequency values that were determined for the term in the subset.

Abstract: Methods, systems, and computer-readable storage media for configuring a request-centric system. Implementations include actions of monitoring operation of the request-centric system to determine a workload of the request-centric system, wherein the workload includes a plurality of requests, determining a mapping function based on a utility function, the workload and a configuration, the mapping function mapping the workload and the configuration into the utility function, determining an optimal configuration based on the mapping function, the optimal configuration allocating resources of the request-centric system to achieve one or more predefined goals, and applying the optimal configuration to the request-centric system.

Abstract: A hierarchical collection of items including one or more sub collections of items ordered in a hierarchy is received. A statistical measure of frequency of an item in a sub collection of items is determined. Further, statistical measures of weightages of the item are determined defining a number of sub collections in the hierarchical collection of items in which the item appears and a number of sub collections in which the item appears. A statistical measure of variability defining a number of occurrences of the item in the hierarchical collection of items across different sub collections is calculated. Furthermore, a relevance score of the item is determined based on the statistical measure of frequency, the one or more statistical measures of weightages of the item and the statistical measure of variability. Based on the relevance score, the item is presented on a computer generated graphical user interface.

Abstract: A method, device and system for maintaining a pre-defined number of free registers at any location of an instrumented program such that at each point during execution of a segment of the instrumented program, the segment has access to the pre-defined number of free registers.

Abstract: In an enhanced file system specialized operators are provided, which consolidate contextually sensitive selections of documents from widely scattered sources in a concise presentation, such as a linear list. One specialized operator, _desc, converts a hierarchical tree into a single level, and provides an exhaustive list of the directories of documents that are distributed throughout the tree. Another specialized operator, _star, provides a single level presentation, such as a linear list, of the child elements of its contextual node. The specialized operators are invoked by opening special directories that are presented to the user in a conventional file system interface.

Abstract: A method, device and system for maintaining a pre-defined number of free registers at any location of an instrumented program such that at each point during execution of a segment of the instrumented program, the segment has access to the pre-defined number of free registers.

Abstract: There are presented a method, device and system for allocating spill cells for an instrumentation fragment that is run on a processor that uses a register stack architecture where only one free register is available for such fragment.

Abstract: A technique for asynchronous computer disk cache management reduces the burstiness of disk I/O. It provides a periodic sync process to flush modified buffers, wherein only buffers exceeding a threshold age are selected to be flushed. In one variant the sync process only executes when the number of modified buffers in the cache exceeds a threshold, which can be dynamically adapted to changing patterns of cache usage. The age of the dirty buffers is measured by sync periods to minimize time computation. In some embodiments an additional cleaning process flushes old buffers between sync invocations.