Abstract: A system for homomorphically generating rotation keys for use in homomorphic computation in association with a client, and a server coupled to the client machine over a network. Client-side code executes in the client to derive a set of Learning With Errors (LWE) ciphertexts from a secret key polynomial, the secret key polynomial having a set of coefficients. Each LWE ciphertext is derived from a coefficient of the secret key polynomial and having a single coefficient. The client-side code transmits the set of LWE ciphertexts to the server. Server-side code receives the set of LWE ciphertexts and processes them into a ring variant (R-LWE) ciphertext homomorphically to generate the one or more rotation keys. The R-LWE ciphertext encrypts a polynomial having the set of coefficients of the secret key polynomial; a given rotation key corresponds to rotated coefficients of that polynomial. The generated rotation keys are useful for FHE computation.

Abstract: An embodiment maps identifying information of a remote registry into a database within a local inventory at a local registry hub. An embodiment selects at least one remote registry from an index maintained in the local inventory in accordance with a policy received at a scheduler from an external client of the local registry hub. An embodiment selects a locally stored image in accordance with a policy received from an external client of the local registry hub. An embodiment uploads replicas of the selected image via one or more registry agents, each registry agent transmitting to its corresponding remote registry, transmitting constituent layers of the replica across multiple remote registries simultaneously such that a subset of the layers constituting the image are uploaded to each remote registry. An embodiment stores metadata for the uploaded image in a cache within a local metadata store.

Abstract: An embodiment senses a raw data sequence by a central processing unit, responsive to the raw data sequence, computes by the central processing unit a transfer data size of the raw data sequence based at least in part on the comparison of a data size of the raw data sequence to a memory size of a graphics processing unit. The embodiment transfers by the central processing unit of the raw data sequence to the graphics processing unit based on the transfer data size. The embodiment trains a foundation model on the raw data sequence where a sliding window algorithm is executed on the raw data sequence by the graphics processing unit, where generating a window of the sliding window algorithm is based on a memory pointer to the raw data sequence.

Abstract: A semiconductor structure for a phase-change memory device includes a heater element on a portion of a bottom electrode in a first dielectric material. The semiconductor structure includes a layer of phase-change material that surrounds a portion of a second dielectric material, where the layer of phase-change material forms a three-dimensional shape around the portion of the second dielectric material. A conductive liner is under a first portion of the layer of phase-change material and surrounds a portion of a bottom surface of a hardmask layer and vertical portions of the hardmask layer. A conductive material is on a portion of a top surface of the second dielectric material and abuts the vertical portions of the layer of phase-change material below the conductive liner and the hardmask layer. A top electrode is on a top surface of the conductive material.

Abstract: Device-implemented methodology for enabling and/or performing crypto-erase via internal action and/or external action in a Key per IO-enabled system. In various approaches, crypto-erasure of the data stored in a Key per IO scheme is enabled by implementing an internal key, which is combined with an external key to generate a media encryption key, which is in turn used to encrypt/decrypt data. By restricting access to the internal key, destruction of the internal key and all media encryption key(s) created using the internal key, renders the data crypto-erased, and thus unrecoverable.

Abstract: In a method for intelligently executing predictive simulator, a processor may input a previous input vector of conditions for a predictive simulator collected at a first time into a machine-learning (ML) model. A processor may input a current input vector of conditions for the predictive simulator collected at a second time into the ML model. A processor may determine using the ML model, a binary similarity index. The binary similarity index represents a prediction of similarity between a first output from the predictive simulator based on the previous input and a second output from the predictive simulator based on the current input.

Abstract: A method, computer system, and a computer program product for smart SDN is provided. The present invention may include recording and clustering a pod's behavior to generate a behavior transition model for the pod. The present invention may include watching a behavior of the pod and comparing the behavior to the generated behavior transition model. The present invention may include triggering a network policy change based on determining that the behavior of the pod is a misbehavior.

Abstract: A signal-analysis-based process for detecting and correcting power feeds is provided. The process includes obtaining voltage waveform data for power feeds connected to a system, and determining for the power feeds, using the obtained voltage waveform data, at least one respective signal characteristic. Further, the process includes comparing the respective signal characteristics of the power feeds to ascertain at least one signal characteristic difference between the power feeds. Based on the at least one signal characteristic difference not exceeding at least one respective difference limit, the process determines that the power feeds are not separately-sourced redundant power feeds connected to the system, and based on determining that the power feeds are not separately-sourced redundant power feeds, the process initiates a corrective action to ensure that the system is connected to separately-sourced redundant power feeds.

Abstract: A computer-implemented system and method operate an automatic and adaptive animal behavioral system. The system receives user configuration and feedback, and generates a training set based on online animal behavior patterns and received user feedback. The system trains one or multiple models based on the training set, if the training set is sufficiently large. It then validates the models, and provides system generated classifications and action types and levels for animal's behavior patterns, if the models' accuracy rate is above a threshold.

Abstract: An embodiment maps identifying information of a remote registry into a database within a local inventory at a local registry hub. An embodiment selects at least one remote registry from an index maintained in the local inventory in accordance with a policy received at a scheduler from an external client of the local registry hub. An embodiment selects a locally stored image in accordance with a policy received from an external client of the local registry hub. An embodiment uploads replicas of the selected image via one or more registry agents, each registry agent transmitting to its corresponding remote registry, transmitting constituent layers of the replica across multiple remote registries simultaneously such that a subset of the layers constituting the image are uploaded to each remote registry. An embodiment stores metadata for the uploaded image in a cache within a local metadata store.

Abstract: Method and apparatus for generating an updated power delivery network. Generating the power delivery network includes determining power characteristics for a power delivery network of a circuit design based on logic cells of the circuit design. The power delivery network includes power wires and power staples connecting pairs of the power wires to each other. Further a first one or more of the power staples is remove from the power delivery network based on the power characteristics. Signal wires for the logic cells are routed after removing the first one or more of the power staples. Routing the signal wires includes routing a first signal wire of the signal wires in a routing track corresponding to the first one or more of the power staples.

Abstract: One or more systems, devices, computer program products and/or computer-implemented methods provided herein relate to determining veracity of answers generated by machine comprehension question and answer models. According to an embodiment, a machine comprehension component can generate a first answer to a query by extracting the first answer from a passage of text corpus. The text corpus alteration component can alter the text corpus one or more times to produce one or more altered text corpora. The machine comprehension component can further extract one or more additional answers to the query from the altered text corpora. A comparison component can determine a veracity score for the first answer based on one or more comparisons of the first answer with the one or more additional answers.

Abstract: According to an embodiment of the present invention, a method, system, and computer program product for preparing a CZ state for use in magic state distillation. The embodiment may include initializing a code state across data qubits. The embodiment may include measuring a CZ operator of the codes state on at least one ancilla qubit proximal to the data qubits. The embodiment may include performing additional quantum operations with the CZ state based on the measurement of the at least one ancilla qubit.

Abstract: A phase-change memory device includes a bottom electrode; a stack of alternating electrical conductor layers directly contacting a top surface of the bottom electrode; a metal pillar directly contacting a top surface of the stack; a phase change material element directly contacting a top surface of the metal pillar; and a top electrode on the phase change material element, wherein a lateral dimension of the metal pillar is smaller than that of the stack.

Abstract: In a method for interpreting output of a machine learning model, a processor receives a first interpretable rule set. A processor may also receive a second interpretable rule set generated from a dataset and model-predicted labels classifying the dataset. A processor may also generate a difference metric and mapping between the first interpretable rule set and the second interpretable rule set.

Abstract: A system and related method reduce public cloud provisioning latencies using one or more processors. The method comprises, prior to receiving a volume provisioning request from a user, creating a pool of pre-provisioned generic volumes. The method further comprises receiving the request from the user to provision a volume from the pool, and then determining that a pre-provisioned generic volume is available for customization based on the request. Responsive to the determination, the method executes the actions comprising customizing the pre-provisioned generic volume based on the request, creating a custom volume, and providing the customized pre-provisioned volume to the user.

Abstract: A data virtualization layer (DV) receives a join query request related to a plurality of tables respectively stored in a plurality of distributed database servers. A plurality of candidate query plans for the join query request is generated where each of the plurality of candidate query plans indicates an order for transmitting the tables respectively stored in the database servers to the DV. For each of the plurality of candidate query plans, a query cost for the candidate query plan is calculated based on a data amount of the tables to be transmitted according to the candidate query plan. From the plurality of candidate query plans, a query plan is determined for the join query request which has a lowest query cost.

Abstract: A computer implemented method includes receiving a list of areas on a subject tape to be read, wherein each area of the list of areas is indicated by a first record number and a last record number corresponding to the area, identifying parameters of a tape drive configured to read the subject tape, wherein the identified parameters of the tape drive contribute to a speed with which the tape drive can read the list of areas, creating a directed graph of the areas on the subject tape based on the identified parameters, wherein the directed graph indicates how long the tape drive will take to read the areas on the subject tape, and determining a fastest reading order of the areas on the subject tape, based, at least in part, on the directed graph and the identified parameters. A computer program product and computer system are also disclosed.

Abstract: A method, system, and computer program product implement a three-factor authorization in a trusted computing environment. The method includes triggering, by a hypervisor, a start of a secure guest by passing control regarding an image of the secure guest and metadata of the secure guest to a trusted firmware, where the secure guest is designed to access a hardware security module (HSM). Upon a successful integrity check of the metadata of the secure guest by the trusted firmware, the secure guest is started using the hypervisor and any sensitive request from the secure guest to the HSM is blocked. The secure guest submits a request with a request structure including a third authorization secret and a characterization of a requested HSM to the trusted firmware. The method also includes binding each HSM protected key generated in the requested HSM in response to the request to the third authorization secret.

Abstract: Embodiments for providing enhanced location-aware protection of latches in a computing environment are provided. One or more latches are combined in one or more of a plurality of bounding boxes on a two-dimensional circuit design layout based on one or more rules. A location-aware interleaving of error correction codes (“ECC”) and burst error correction codes may be selectively applied to one or more latches in those of the plurality of bounding boxes, where multiple bit errors are corrected.

Abstract: A method, system, computer program product and computer program for managing a storage system, the storage system comprising a first storage device, a second storage device, a source storage device, a target storage device, and a cache, the method comprising: receiving a first host access request for a data block on the first storage device; receiving a mapping between the source storage device and the target storage device; and in response to the first host access request and the mapping: accessing a corresponding data block on the second storage device; and updating metadata for the cache.

Abstract: A system, computer program product, and a computer implemented method are provided for interfacing with a virtual dialog environment to dynamically and optimally collected context for problem diagnosis and resolution. A context model is leveraged to identify context entities, and one or more corresponding context collection mechanisms. The context model is implemented in real-time to facilitate dynamic selection of one or more of the identified context collection mechanisms, which are selectively subject to execution to resolve the problem diagnosis.

Abstract: Described are techniques for application hardening. The techniques include generating application traces using fuzzing for an application with a known security vulnerability, where the application traces include good traces that do not result in exploitation of the known security vulnerability and bad traces that result in exploitation of the known security vulnerability. The techniques further include identifying code segments that are executed by the bad traces and not executed by the good traces. The techniques further include modifying the identified code segments using binary rewriting.

Abstract: A 2D microfluidic structure for capacitance sensing of analyte is provided. The structure includes a first substrate located above at least one microfluidic channel, and a second substrate located below the at least one microfluidic channel. The first substrate includes at least one first group of three isolated electrodes and the second substrate includes at least one second group of three isolated electrodes, where each group of isolated electrodes includes a ground electrode and two probe electrodes.

Abstract: Aspects of this disclosure include a recall notification method, a method for communicating an action to be performed by a device, and a method of operating a relay service by a network device. One embodiment of the recall notification method may comprise periodically transmitting, by a network interface of a medical device implanted in a user's body, a poll request to a recall service endpoint. The method may further comprise receiving, by the network interface in response to the poll request, a recall notification; and in response to receiving the recall notification, directly notifying the user.

Abstract: A computer implemented method for merging loops. A number of processor units identifies loops in computer code. The loops are sequences of instructions that are repeated until conditions for the loops are reached. The number of processor units creates a tree comprising nodes that represent the loops and edges that represent relationships between nodes. The number of processor units utilizes the tree to identify a pair of candidate loops from sibling nodes. The number of processor units creates a new loop from the pair of candidate loops with an expanded iteration space based on iteration spaces for the pair of candidate loops in response to the pair of candidate loops being eligible for merging.

Abstract: A method for system authentication includes subjecting a system to a challenge. The method further includes receiving a response from the system. The received response is dependent upon a location of a filament in a resistive random-access memory device of the system. Additionally, the response is also a unique identifier.

Abstract: A first operation in a computing environment with a set of nodes is monitored. A subset of nodes that are utilized by the first operation is identified. The subset of nodes is a subset of the set of nodes. For each node in the subset of nodes, an operation counter for the respective node is incremented and an operation time for the first operation is added to a total operation timer for the respective node. For each node in the subset of nodes, the respective total operation timer is divided by the respective operation counter. A node in the subset of nodes that appears to be slow is identified based on the dividing. The data for multiple operations could be aggregated in the operation counters and total operation timers for each node within the set of nodes, enabling nodes that appear to be slow to be identified on aggregate data.

Abstract: A plurality of different images of a same region of interest in an object are input into a set of neural networks, wherein each image of the region has been captured under a different value of a variable condition. A classification for each image is generated by the set of neural networks, wherein each classification includes a confidence score in a prediction of whether a feature is present in the region. The image classifications are ensembled to generate a final classification for the region. By applying a loss function, a loss is computed based on comparing the final classification to a ground truth of whether the feature is present in the region. The parameters of the set of neural networks are adjusted based on the computed loss.

Abstract: Described are techniques for disintegrating an entity into smaller entities. A graph (“first graph”) for the entity of records to be disintegrated is constructed, where each vertex of the first graph represents a record in the entity of records to be disintegrated. The edges in the first graph connecting records in the entity of records represent matching links between the records, where each edge is associated with a weight corresponding to a similarity score. Furthermore, two or more additional graphs representing two or more sub-entities of the entity of records to be disintegrated are constructed. Such graphs are constructed based on selecting edges with a maximum weight out of the edges connected between each pair of records in the first graph or based on the number of connections each record has with other records in the first graph exceeding a threshold value.

Abstract: In an approach for building an automated customer support system, a processor receives a set of sentences extracted from a natural language conversation occurring between an IT support system and a user. A processor extracts an initial state and a goal state from the set of sentences using a Natural Language Classifier. A processor extracts one or more actions from the set of sentences. A processor creates a formal planning model. A processor determines the one or more formal actions are not complete using a first machine learning model. A processor completes the one or more formal actions with one or more missing parts. A processor produces an executable plan using a planner. A processor implements one or more executable scripts according to a sequence of the one or more formal actions of the executable plan using a plan executor.

Abstract: A magnetic random access memory (MRAM) stack, a method of fabricating a MRAM stack, a MRAM array, and a computer system. The MRAM stack includes a first magnetic layer comprising a Heusler compound. The MRAM stack also includes one or more seed layers including a templating structure having a crystalline structure configured to template the Heusler compound. The magnetic layer is formed over the templating structure. The MRAM stack also includes a chromium (Cr) layer formed under the templating structure. The Cr layer is configured to enhance a tunnel magnetoresistance (TMR) of the MRAM stack.

Abstract: A method, system, and computer program product for frictionless mutual authentication of unsolicited communications may detect an incoming communication. A verification interface may be displayed on a consumer device. On the consumer device, a first valid verification may be received via the verification interface. In response to receiving the first valid verification, a challenge interface may be presented to an enterprise device. On the enterprise device, a second valid verification may be received via the challenge interface. In response to receiving the second valid verification, a verification credential may be presented to both the consumer device and the enterprise device. A connection for the incoming communication may be established between the consumer device and the enterprise device.

Abstract: Embodiments of the present invention provide methods, computer program products, and systems. Embodiments of the present invention can condense a hierarchy in a data governance system, wherein the hierarchy comprises a root node and at least one child node comprising related sub-trees by determining, for a parent node in the hierarchy of governance system, governance terms and respective assignment relationships from a plurality of information assets, determining usage of the governance term in at least one of a plurality of governance rules, and marking a governance term of the plurality of governance terms for elimination based on the determined assignment relationships and the determined usage of the governance term in the plurality of governance rules. Embodiments of the present invention can then delete the governance term from the hierarchy if the governance term is marked for elimination.

Abstract: Obtain, as input, in electronic form, structured information including tasks for a plurality of occupations in a plurality of industries over a length of time; compute, from the structured information, a time series of normalized occupation task shares over the length of time; train a computerized machine learning model, on the time series, to predict future task shares for the plurality of occupations in the plurality of industries; and, with the trained computerized machine learning model, predict the future task shares.

Abstract: A machine learning model can be run with a first input data to obtain a first output from the machine learning model. A first rule representing an interpretable explanation for the machine learning model predicting the first output can be obtained. A modified rule can be received, which changes the first rule. A transformation function can be generated, which transforms a given input data for inputting to the machine learning model to a new input data based on decision boundaries of the machine learning model and the modified rule. The first rule, the first output, the modified rule, and the transformation function can be stored in a first table.

Abstract: A method protects a computer asset by identifying a particular signature, which is software that causes a particular gateway to block an intrusion from reaching a particular computer asset, and installs the particular signature on the particular gateway, thus protecting the computer asset from the intrusion.

Abstract: A system can compensate for activation drift in analog memory-based artificial neural networks. A set of input activation vectors can be input, at a first point in time, to a crossbar array. The first set of output activation vectors can be read from the output lines of the crossbar array. At a second point in time, which is a later time than the first point in time, the input set of activation vectors can be input to the crossbar array. A second set of output activation vectors can be read from the crossbar array. A function that maps the second set of output activation vectors to the first set of output activation vectors can be determined. The function can be applied to subsequent output activation vectors output by the crossbar array. A method thereof, can also be provided.

Abstract: An optical cable mismatch error detection and remediation process is provided which includes detecting, by at least one processor, an optical cable mismatch error in a connection path between optical transceivers of a computing network, where the connection path includes one or more optical cables. The detecting includes evaluating operation of an optical transceiver of the connection path. The evaluating includes determining that data on one or more optical lanes of the optical transceiver is lane-mismatched data. The process further includes restructuring one or more lane assignments of the optical transceiver, based on determining that the data on the one or more optical lanes of the optical transceiver is lane-mismatched data, where the restructuring addresses the detected optical cable mismatch error.

Abstract: A semiconductor device includes a plurality of field effect transistors (FET) formed upon semiconductor fins. Each FET includes a gate disposed transversely upon a first portion of the fins of the FET, one or more source/drain regions disposed upon the fins and in contact with the gate, and an electrically isolating layer disposed adjacent to a second portion of the fins above the gate and the source/drain regions, the electrically isolating layer having an interface with the gate. The device further includes a buried power rail (BPR) disposed between otherwise adjacent FETs. The BPR includes a metal rail extending beyond the interface into the gate, and electrically isolating sidewalls separating the metal rail from the gate and the source/drain regions. The device also includes a via-buried power rail contact disposed adjacent to the electrically isolating sidewalls, in contact with the metal rail, and in contact with one source/drain region.

Abstract: A novel bevel etch sequence for embedded metal contamination removal from BEOL wafers is provided. In one aspect, a method of processing a wafer includes: performing a bevel dry etch to break up layers of contaminants with embedded metals which, post back-end-of line metallization, are deposited on a bevel of the wafer, which forms a damaged layer on surfaces of the wafer, and then performing a sequence of wet etches, following the bevel dry etch, to render the bevel of the wafer substantially free of contaminants, wherein the sequence of wet etches includes etching the damaged layer to undercut and lift-off any remaining contaminants. A wafer, processed in this manner, having a bevel that is substantially free of contaminants is also provided.

Abstract: In an approach to improve risk assessment and management, embodiments monitor a predetermined area by collecting, by one or more sensing devices, data related to environmental conditions and spatial relationships between one or more objects in the predetermined area. Further, embodiments determine, based on a model and the monitoring, that an accidental event is likely to occur within a predetermined distance or within the predetermined area. Additionally, embodiments identify a predetermined mitigation action from a knowledge corpus to apply to the accidental event; and proactively implementing the identified mitigation action to prevent or manage the accidental event.

Abstract: A set of stacked transistors, system, and method to connect the gates of stacked field-effect transistors through sidewall straps. The set of stacked transistors may include a first transistor including a first gate. The set of stacked transistors may also include a second transistor including a second gate, where the second transistor is above the first transistor. The set of stacked transistors may also include a dielectric preventing direct contact between the first gate and the second gate. The set of stacked transistors may also include a first sidewall strap proximately connected to the first gate and the second gate, where the first sidewall strap connects the first transistor and the second transistor.

Abstract: A fastener system includes a bar, a knob, a bushing, and a washer. The bar extends along an axis and includes a catch at one end and a head at the other end, where the head includes an inclined face. The knob is configured to engage with the catch to prevent relative rotation between the knob and the bar and to prevent separation of the knob and the bar along the axis. The bushing is comprised of an elastomeric material that is positioned around the bar between the catch and the head, where the bushing includes a flange facing the knob. The washer is positioned around the bar between the bushing and the head, where the washer includes an inclined face that contacts the other inclined face. The coefficient of friction between the bushing and the washer is higher than the coefficient of friction between the washer and the head.

Abstract: The present disclosure relates to a method comprising receiving a request to execute a workload using an artificial intelligence model. A current resource utilization status in the distributed system may be determined. The current resource utilization status may be used to define a deployment configuration of the artificial intelligence model, wherein the deployment configuration is defined by: a number and structure of input blocks, a number and structure of output blocks and the intermediate block of the artificial intelligence model, a second computer system to execute the intermediate block, and one or more first computer systems to execute the input and output blocks. The artificial intelligence model may be deployed in accordance with the defined deployment configuration and the workload may be executed.

Abstract: Altering an automated conversation system by receiving automatic conversation log data, generating a decision tree from the data, summarizing child node data for a node of the decision tree, generating node response variations according to the child node data of the node, and altering the automatic conversation according to the node response variations.

Abstract: A computer-implemented method is provided for policy evaluation. In the method, the utility of the given decision-making policy is estimated based on a dataset of state-action-reward-state tuples, a set of candidate bootstrapping estimators of the fitted Q-evaluation (FQE) algorithm, and a criterion function. The method automatically selects the best bootstrapping estimator from the candidates based on the criterion function and, when the criterion function is appropriately designed, it produces a good policy-value estimate such that the estimation error is small (below a threshold).

Abstract: Described are techniques for automated labeling of cybersecurity incidents. The techniques include generating a set of labels for a received cybersecurity incident based on features of the received cybersecurity incident. The techniques further include prioritizing the set of labels to generate a subset of labels, and associating the subset of labels to the received cybersecurity incident.

Abstract: Embodiments of the invention are directed to a system for detecting neurotransmitters. A non-limiting example of the system includes a porous electrode. A system can also include a pH sensor attached to the porous electrode, wherein the pH sensor includes a sensing electrode and a reference electrode. The system can also include electronic circuitry in communication with the pH sensor.

Abstract: An approach for intelligent optimization of machine learning models for a target environment may be provided herein. The approach may include extracting metadata from a training pipeline for a first machine learning model that has been configured to operate within a first computing environment. The approach may also include mapping the extracted metadata to one or more constraints associated with a second machine learning model that has been configured to operate within a second computing environment. The approach may also include training the second machine learning model configured to the second computing environment, with the dataset that was used to train the first machine learning model, based on the mapped constraints. The approach may also include comparing performance metrics of the first machine learning model to corresponding metrics of the now trained second machine learning model.