Abstract: The present invention relates to the synthesis and production of biopolymer from natural waste—the husk of coconut and/or coir pith powder. More specifically it pertains to a novel multi-functional biopolymer obtained from the husk of coconut and/or coir pith powder, the fruit of Cocos nucifera with combinations of copolymer, crosslinker, and plasticizers for use in pharmaceutical, nutraceutical, cosmeceutical and leather industry. Lignin component from coconut husk was extracted, purified, studied for their physicochemical properties and processed into various biopolymers to be used as burn wounds dressings and leather substitutes.

Abstract: Systems and methods for digital audio measurement such as perceived loudness. An example of the method includes dividing, by a processor, a digital audio file into a plurality of blocks in a time sequence at a first time length; determining, by the processor, a respective Loudness Units relative to Full Scale (LUFS) for each block; determining, by the processor, a difference between the LUFS of each block with a reference value; and generating, by the processor, an indicator from one or more of the differences of the plurality of blocks within a second time length.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors for distributed training of a neural network. One of the methods includes receiving, at each of the plurality of devices, a respective batch; performing, by each device, a forward pass comprising, for each batch normalization layer: generating, by each of the devices, a respective output of the corresponding other layer for each training example in the batch, determining, by each of the devices, a per-replica mean and a per-replica variance; determining, for each sub-group, a distributed mean and a distributed variance from the per-replica means and the per-replica variances for the devices in the sub-group; and applying, by each device, batch normalization to the respective outputs of the corresponding other layer generated by the device using the distributed mean and the distributed variance for the sub-group to which the device belongs.

Abstract: A mounting and alignment system for building panels. A mullion guides one edge of a first panel mounted on a structure interacts with a mullion guide on an adjacent edge of second panel to be mounted next to the first. The mullion guides engage and interact as the second panel is lowed in place to adjust the horizontal position of the second panel relative to the first to a small tolerance suitable for other interacting panel structures. After the second panel is installed, mullion guides on the adjacent panel edges can be removed.

Abstract: A mounting and alignment system for building panels. A mounting bracket comprises a lower component affixable to a building support and an upper component from which a panel can be hung. A bearing assembly between the upper and lower components transfers load from the upper component to the lower component and allows the horizontal position of the upper component and thereby the panel to be easily adjusted. Mullion guides on the sides and extending over the panel top interact with mullion guides on other panels and automatically adjust the horizontal position of a panel as the panel is lowed in place from a relatively large horizontal placement tolerance a smaller placement tolerance required for other interacting structures on adjacent sides of installed panel. After the panel is installed, mullion guides on the adjacent panel sides can be removed.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors for distributed training of a neural network. One of the methods includes receiving, at each of the plurality of devices, a respective batch; performing, by each device, a forward pass comprising, for each batch normalization layer: generating, by each of the devices, a respective output of the corresponding other layer for each training example in the batch, determining, by each of the devices, a per-replica mean and a per-replica variance; determining, for each sub-group, a distributed mean and a distributed variance from the per-replica means and the per-replica variances for the devices in the sub-group; and applying, by each device, batch normalization to the respective outputs of the corresponding other layer generated by the device using the distributed mean and the distributed variance for the sub-group to which the device belongs.

Abstract: A mounting and alignment system for building panels. A mounting bracket comprises a lower component affixable to a building support and an upper component from which a panel can be hung. A bearing assembly between the upper and lower components transfers load from the upper component to the lower component and allows the horizontal position of the upper component and thereby the panel to be easily adjusted. Mullion guides on the sides and extending over the panel top interact with mullion guides on other panels and automatically adjust the horizontal position of a panel as the panel is lowed in place from a relatively large horizontal placement tolerance a smaller placement tolerance required for other interacting structures on adjacent sides of installed panel. After the panel is installed, mullion guides on the adjacent panel sides can be removed.

Abstract: Techniques for enhanced Software-Defined Wide Area Network (SD-WAN) path quality measurement and selection are disclosed. In some embodiments, a system/method/computer program product for enhanced SD-WAN path quality measurement and selection includes periodically performing a network path measurement for each of a plurality of network paths at a Software-Defined Wide Area Network (SD-WAN) interface; updating a version if the network path measurement exceeds a threshold for one or more of the plurality of network paths; and selecting one of the plurality of network paths for a session based on the version according to an application policy.

Abstract: Certain aspects of the present disclosure provide techniques for managing the lifecycle of user data. A user data lifecycle management system can collect user data from multiple sources including the user, the organization implementing the lifecycle management system, and third parties. The user data is used to create a user profile with a global user identifier. The user profile is scored based on the attributes within the user profile. The user data lifecycle management system can route the user profile to a destination source based on the score of the user profile. The destination source can be a tool of the organization for interacting with the user. The destination sources can also provide feedback to that is incorporated to the user profile and can assist the user data lifecycle management system in managing the user profile and aid in decision making.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors for a network having one or more degraded nodes. A method comprises training a respective replica of a machine learning model on each node of multiple nodes organized in an n-dimensional network topology, combining the respective individual gradient vectors in the nodes to generate a final gradient vector by performing operations comprising: designating each group of nodes along the dimension as either a forwarding group or a critical group, updating, for each receiving node, a respective individual gradient vector with an intermediate gradient vector, performing a reduction on each critical group of nodes along the dimension to generate a respective partial final gradient vector for the critical group, and updating, for each critical group of nodes, an individual gradient vector for a representative node with the respective partial final gradient vector.

Abstract: Systems and methods include reception of an indication of a first data object to transmit from a first database tenant to a second database tenant, the first data object comprising user configuration settings metadata of a database application and associated with a first data object type, determination of whether a dependent data object may be associated with a data object of the first data object type, determination, if it is determined that a dependent data object may be associated with a data object of the first data object type, of whether a dependent object is dependent on the first data object, and automatic determination, if it is determined that a dependent object is dependent on the first data object, to transmit the dependent object from the first database tenant to the second database tenant.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors and similarly structured data that are generated in parallel, for example, on nodes organized in a mesh or torus topology defined by connections in at least two dimension between the nodes. The methods provide parallel computation and communication between nodes in the topology.

Abstract: Generating a new disaster recovery workflow is provided. In response to determining that a failed action was detected during execution of a disaster recovery workflow, reasons and fixes corresponding to the failed action are acquired from a data source. A set of correlated corrective actions that are potential fixes for the failed action is identified based on natural language processing of the reasons and fixes corresponding to the failed action. A weightage value is assigned to each correlated corrective action in the set of correlated corrective actions based on a plurality of factors to form a set of corrective actions with weightage values. A recommended new disaster recovery workflow is generated by embedding the set of corrective actions with weightage values within the disaster recovery workflow.

Abstract: Examples for enhancing veracity of data are described herein. Data from a repository may be received based on a data receiving rule. From the received data, a first dataset may be generated using statistical modeling. Also, a first data veracity score for the first dataset is generated which is indicative of a degree of usability of the dataset. Another aspect relates to identifying an anomaly in the first dataset, the corrector, for each anomaly, to identify a correction technique from amongst a plurality of correction techniques. Further, a second dataset is generated using the identified correction technique having second data veracity score higher than the first data veracity score.

Abstract: Techniques for enhanced Software-Defined Wide Area Network (SD-WAN) path quality measurement and selection are disclosed. In some embodiments, a system/method/computer program product for enhanced SD-WAN path quality measurement and selection includes periodically performing a network path measurement for each of a plurality of network paths at a Software-Defined Wide Area Network (SD-WAN) interface; updating a version if the network path measurement exceeds a threshold for one or more of the plurality of network paths; and selecting one of the plurality of network paths for a session based on the version according to an application policy.

Abstract: Examples for enhancing veracity of data are described herein. Data from a repository may be received based on a data receiving rule. From the received data, a first dataset may be generated using statistical modeling. Also, a first data veracity score for the first dataset is generated which is indicative of a degree of usability of the dataset. Another aspect relates to identifying an anomaly in the first dataset, the corrector, for each anomaly, to identify a correction technique from amongst a plurality of correction techniques. Further, a second dataset is generated using the identified correction technique having second data veracity score higher than the first data veracity score.

Abstract: Techniques for enhanced Software-Defined Wide Area Network (SD-WAN) path quality measurement and selection are disclosed. In some embodiments, a system/method/computer program product for enhanced SD-WAN path quality measurement and selection includes periodically performing a network path measurement for each of a plurality of network paths at a Software-Defined Wide Area Network (SD-WAN) interface; updating a version if the network path measurement exceeds a threshold for one or more of the plurality of network paths; and selecting one of the plurality of network paths for a session based on the version according to an application policy.

Abstract: Generating a new disaster recovery workflow is provided. In response to determining that a failed action was detected during execution of a disaster recovery workflow, reasons and fixes corresponding to the failed action are acquired from a data source. A set of correlated corrective actions that are potential fixes for the failed action is identified based on natural language processing of the reasons and fixes corresponding to the failed action. A weightage value is assigned to each correlated corrective action in the set of correlated corrective actions based on a plurality of factors to form a set of corrective actions with weightage values. A recommended new disaster recovery workflow is generated by embedding the set of corrective actions with weightage values within the disaster recovery workflow.

Abstract: The present disclosure relates generally to an awareness enhancement apparatus and method for undesirable repeated behaviors, including but not limited to obsessive compulsive and related disorders, and most relevant to trichotillomania (hair pulling), onychophagia (nail biting), dermatillomania (skin picking), and thumb sucking, among others. More particularly, the disclosure relates to a sensing and feedback device and associated methods of use which indicate a behavior based on the user's physical gestures and positioning of the hands, these gestures and positions being related to these undesirable behaviors typical of such disorders, and alerting the user so that he or she can reduce the behavior.

Abstract: In a L×M×N 3D torus network of computer nodes, a 2-dimensional plane comprising M×N torus network of nodes mapped into M/2 meshes×2*N torus network of nodes. N can be k*M, k is an integer greater than zero, and M and N are even numbers. Each of M/2 mesh of the 2*N torus is contiguous in the 2D plane. Mapping is performed for each of the L planes of the L×M×N 3D torus network. The M/2 meshes are combined with a remaining torus network dimension comprising L planes, the combining creating another 2*N pattern, wherein an L*M/2×2*N communication pattern is created. Application entities are executed according to the mapped L*M/2×2*N communication pattern.