Abstract: In one general embodiment, an apparatus includes a magnetic head, and a controller coupled to the magnetic head, the controller being configured to perform a method that includes measuring humidity in a vicinity of the apparatus in response to detection of an error while performing a data operation on a magnetic recording tape. An error recovery procedure to remove debris on the tape is executed in response to detection of an error and the measured humidity exceeding a predefined upper limit value. An error recovery procedure to remove debris adhering to the magnetic head is executed in response to detection of an error and the measured humidity being below a predefined lower limit value.

Abstract: An item storage system includes a shelving portion and a replenishment portion. The shelving portion includes: a first shelf having a toroidal shape with a center axis; and a second shelf having a toroidal shape that is spaced apart from the first shelf along the center axis. The replenishment portion includes: an elevator portion configured to deliver items to the first shelf and the second shelf, wherein at least part of the elevator portion is positioned in the center of the shelving portion and is surrounded by the shelving portion; and a storing portion configured to load items into the elevator portion, wherein the storing portion is positioned above or below the shelving portion.

Abstract: A method for laser assisted delivery of therapeutic agents includes selectively controlling a valve connected to a first channel disposed within a first sidewall of a nozzle, applying, through the first channel, a first substance to penetrate dermis to a predetermined depth, and administering, through a second channel unconnected with the valve and disposed within a second sidewall of the nozzle, a second substance to remove debris.

Abstract: An exemplary method includes forming a sacrificial layer along sidewalls of an array of trenches that are indented into a substrate, depositing a fill layer over the sacrificial layer, and then creating an array of gaps between the fill layer and the substrate by removing the sacrificial layer along the sidewalls of the trenches, while maintaining a structural connection between the substrate and the fill layer at the floors of the trenches. The method further includes covering the substrate, the fill layer, and the gaps with a cap layer that seal fluid-tight against the substrate and the fill layer. The method further includes indenting a first reservoir and a second reservoir through the cap layer, and into the substrate and the fill layer, across the lengths of the array of gaps, so that the array of gaps connects the first reservoir in fluid communication with the second reservoir.

Abstract: An efficient pipelined implementation of digital scaling, offset and aggregation operation supports element-by-element programmable scale and offset factors. The method includes time-multiplexed parallel pipelining of a plurality of digital data words, each of the plurality of digital data words encoding an N-bit signed integer, from one of a plurality of receive-registers through a datapath that can either (1) store the plurality of digital data words directly in a dedicated first memory, (2) store the plurality of digital data words directly in a dedicated second memory, or (3) direct the plurality of digital data words into a parallel set of fused-multiply-add units. The method further includes multiplying each digital data word by a corresponding data-word retrieved from the dedicated first memory to form product data words and adding the product data words to a corresponding data-word retrieved from the dedicated second memory to form an output sum-and-product data words.

Abstract: Embodiments of the present invention manage a state of charge of a rechargeable battery for extended storage by determining a manual override for a storage protocol is not activate for a rechargeable battery associated with a battery charger and an electronic device. Receiving battery data, environment data, and historical data for the rechargeable battery associated with a battery charger. Embodiments of the present invention determine to activate the storage protocol for the rechargeable battery based on the battery data, the environment data, and the historical data and discharge the rechargeable battery to a preset state of charge level based on the storage protocol.

Abstract: Generating soft labels used for training a unified model is achieved by unification of models having respective target classes with distillation. A collection of samples is prepared. Predictions are generated by individual trained models. Individual trained models have an individual class set to form a unified class set that includes target classes. The unified soft labels are estimated for each sample over the target classes in the unified class set from the predictions using a relation connecting a first output of each individual trained model and a second output of the unified model. The unified soft labels are output to train a unified model having the unified class set.

Abstract: An example operation may include one or more of receiving, via a custodial service, a request to transact with a digital asset owned by a user and temporarily in custody of the custodial service, generating a blockchain transaction comprising an identifier of the digital asset on a blockchain ledger, an identifier of the custodial service, and an identifier of a recipient of the digital asset, signing the blockchain transaction with a key from a trifocal key which proves that the custodial service is authorized to transact with the digital asset on behalf of the user, and storing the signed blockchain transaction on a blockchain ledger.

Abstract: The method provides for receiving a plurality of documents including mentions of a target entity from a search query about the entity. The mentions of the target entity are identified in respective documents of the plurality of documents. Content surrounding the one or more mentions of the target entity are extracted with the mentions within the respective documents and form section. A respective document includes a plurality of sections. Metrics of relevance and irrelevance to the target entity are determined within the plurality of sections of the respective documents. A density score is generated for the plurality of sections of the respective documents. A relevancy score is assigned to respective documents of the plurality of documents, based on the density scores of the sections of the respective documents. The documents are ranked based on the relevancy score and presented in an order based on the ranking.

Abstract: Aspects of the present invention disclose a method, computer program product, and system for stale data recovery using virtual storage metadata. The method includes one or more processors generating a primary virtual storage metadata structure having virtual address areas for a data chunk. Each area includes a plurality of most recent updates of metadata for the data chunk. The metadata of each update includes a sequence number for the update and a pointer to the physical location of the data. The method further includes one or more processors generating an overflow virtual storage metadata structure for multiple virtual address areas. The overflow metadata structure includes updates that have overflowed from the virtual address areas. In addition, wherein an oldest update in a full virtual address area of the primary virtual storage metadata structure includes a link to an overflow location.

Abstract: A serverless computing-based, continuous gateway watch of a data store for data change process is provided. The process includes the gateway interface of the computing environment receiving a watch request from a user system to monitor the data store for data change. Based on receiving the watch request, the gateway interface invokes a serverless setup service to establish a connection between the gateway interface and the data store of the computing environment to be monitored for data change. Based on receiving, at the gateway interface, a data change indication from the data store, the gateway interface invokes a serverless message process service to mutate the data change indication from the data store into a mutated data change message indicative of a data change at the data store for return to the user system pursuant to the watch request, with the serverless message process service terminating thereafter.

Abstract: From transaction data, a graph modeling a set of relationships between transaction elements is constructed, a node of the graph representing a transaction element, an edge of the graph representing a relationship between two transaction elements. Using a part of speech analysis model, a subset of the nodes is tagged with a corresponding first part of speech tag. An alternative part of speech tag is generated for a tagged node. A set of tagged nodes is grouped into a word group including at least one tagged node having an alternative part of speech tag. The word group is validated against a grammar of a natural language. Responsive to the validating determining that the word group conforms to the grammar, a set of additional nodes representing the word group is added to the graph. Using the modified graph, a transaction in the transaction data is validated.

Abstract: Methods, computer program products, and/or systems are provided that perform the following operations: identifying a first workload being processed by a first plurality of sites in a cluster; identifying, from the first plurality of sites: (i) a first site as a primary site for the first workload, and (ii) one or more secondary sites for the first workload; identifying a communication link issue between the first site and at least one of the one or more secondary sites; splitting the cluster into sub-clusters based, at least in part, on the communication link issue, wherein the first site is included in a first sub-cluster of the sub-clusters and the at least one of the one or more secondary sites is included in a sub-cluster of the sub-clusters that is different from the first sub-cluster; and instructing the first sub-cluster to locally process the first workload.

Abstract: A system includes a virtual reality system configured to enable a user to interact with a virtual environment, a plurality of biofeedback sensors configured to monitor a user, and a computer system including a virtual reality module configured to generate at least a view of the virtual environment, a biofeedback module configured to fuse output of the plurality of biofeedback sensors with a plurality of events within the virtual environment, a training module configured to generate a model of user behavior, wherein the training module executed by the computer system enables the computer system to make a prediction of a user response of the user based on a corpus of biofeedback data, and an alert module configured to generate at least one alert to the user via the virtual reality system based on the user response predicted by the computer system.

Abstract: A system and method for generating hard training data from easy training data. Training data including visual data with synthetic semantic implants (“VSSI”) having at least one cue is received. An annotator identifies at least one cue in the VSSI and annotates the VSSI to indicate the cue to create a modified training data set. A data scrambler removes at least one cue from the VSSI to create the tagged training data, which can then be used to train a classifier to identify transitions between segments when the cues are not present.

Abstract: One or more systems, devices, computer program products and/or computer-implemented methods of use provided herein relate to regularizing semantic similarity relationships relative to a pair of languages. A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise a computation component that generates a transformation comprising a semantic similarity relationship between detected semantic labels of a pair of languages.

Abstract: According to one embodiment, a method, computer system, and computer program product for synchronizing a physical environment with a virtual environment is provided. The present invention may include analyzing a virtual environment to determine one or more physical movements required for a user to navigate within the virtual environment; scanning, with one or more sensors, a physical environment to identify one or more physical movements required for the user to navigate within the physical environment; responsive to determining that the one or more physical movements required for a user to navigate within the virtual environment do not match the one or more physical movements required for a user to navigate within the physical environment, identifying how the physical environment must be modified to incorporate the physical movements of the virtual environment; and communicating the identified modifications to the physical environment to the user.

Abstract: A computer-implemented method for software detection is disclosed. The computer-implemented method includes scanning a list of file systems present on a computer system and described by a respective mount point for a signature from a set of predetermined signatures. The computer-implemented method further includes responsive to detecting the signature from the set of predetermined signatures, selecting a particular plugin from a predetermined list of plugins based, at least in part, on the detected signature. The computer-implemented method further includes querying, using the plug-in, an operating environment of the filesystem for data indicative of software running in the operating environment. The computer-implemented method further includes determining software running in the operating environment based, at least in part, on the data returned from querying the operating environment of the filesystem.

Abstract: In a computer-implemented method for improving performance of a database optimizer, a processor calculates a first execution time for a database query executing through a first access path of a database system. A processor generates a second access path. A processor determines a second execution time via a simulation of the database query through the second access path. A processor determines that the second execution time is less than the first execution time. A processor, responsive to determining that the second execution time is less than the first execution time, sends the second access path to the database optimizer.

Abstract: A method, apparatus and computer program product to protect a deep neural network (DNN) having a plurality of layers including one or more intermediate layers. In this approach, a training data set is received. During training of the DNN using the received training data set, a representation of activations associated with an intermediate layer is recorded. For at least one or more of the representations, a separate classifier (model) is trained. The classifiers, collectively, are used to train an outlier detection model. Following training, the outliner detection model is used to detect an adversarial input on the deep neural network. The outlier detection model generates a prediction, and an indicator whether a given input is the adversarial input. According to a further aspect, an action is taken to protect a deployed system associated with the DNN in response to detection of the adversary input.