Abstract: A computer-implemented method, system and computer program product for performing real-time monitoring of machine learning models. Real-time model state data and metadata (e.g., operating dataset) of the machine learning models located within an orchestration plane of a network are collected by agents located within the machine learning models. The portion of the collected real-time model state data and metadata that is to be provided to the user by the service orchestrator (configured to monitor the machine learning models in the service orchestration plane via the use of agents in the machine learning models) is selected and marked. The marked collected real-time model state data and metadata are then provided to the user by the service orchestrator. In this manner, real-time monitoring of the machine learning models in the orchestration plane, such as the service orchestration plane, of a broadband cellular network (e.g., fifth generation broadband cellular network) is achieved.

Abstract: Embodiments disclosed herein relate to enabling direct wireless device-to-device communication between sleepy end devices (SEDs) of a mesh (e.g., Thread®) network. A router may forward packets between end devices of the mesh network. However, if the router is not available, SEDs may not be able to communicate with each other using a mesh protocol. Embodiments presented herein enable end devices of the mesh network to communicate directly, without a router. Some embodiments are directed to changing a role of an end device to temporarily act as a router for a particular target end device. The role change may be based on a trigger event and may be temporary until a target action is performed by the target end device. In some embodiments, the end devices continue to operate as SEDs and use coordinated sampled listening techniques to communicate via the mesh protocol.

Abstract: A system, method, and computer program product for implementing dynamic telecom network agent filtering is provided. The method includes retrieving peripheral data associated with user equipment (UE) enabled with respect to a telecom network associated with a telecom network hardware device. The peripheral data is mapped with a pre-defined filter selection model and a filtering agent model is selected. Likewise, filtering agents associated with the UE are selected and environmental characteristics associated with the telecom network are retrieved. A location of the UE is tagged and the filtering agents are pushed to the UE. Network commands associated with execution of the filtering agents are generated and the filtering agents are executed with respect to the UE. In response, noise and distortion occurring during operation of the UE with respect to the telecom network is reduced.

Abstract: A method, a computer system, and a computer program product for tiering migration for storage system optimization is provided. Embodiments of the present invention may include determining to move data based on access counters to balance a workload between storage drives and storage tiers. Embodiments of the present invention may include establishing a connection with an interconnect application program interface. Embodiments of the present invention may include collecting and converting physical access counters. Embodiments of the present invention may include collecting heat map data. Embodiments of the present invention may include calculating a tiering migration plan for an internal map change event.

Abstract: A computer-implemented method for altering an appearance of an electrochromic coating of a vehicle is provided. The method includes determining that a vehicle is a boundary area of a testing zone based on a determined location of the vehicle. The method also includes changing a base appearance of the electrochromic coating of the vehicle to an altered appearance based on the determination that the vehicle is in the boundary area.

Abstract: Systems and methods that offer significant improvements to current chatbot conversational experiences are disclosed. The proposed systems and methods are configured to manage conversations in real-time with human customers based on a dynamic and unscripted conversation flow with a virtual assistant. In one embodiment, a knowledge graph or domain model represents the sole or primary source of information for the virtual assistant, thereby removing the reliance on any form of conversational modelling. Based on the information provided by the knowledge graph, the virtual agent chatbot will be equipped to answer customer queries, as well as demonstrate reasoning, offering customers a more natural and efficacious dialogue experience.

Abstract: Embodiments relate to a front-end scaler circuit configured to receive and process demosaiced image data in different modes depending on if the demosaiced image data was demosaiced from Bayer or Quad Bayer raw image data. The front-end scaler circuit shares memory with a demosaicing circuit, and is configured to perform different operations that use different amounts of the shared memory based on the original image format of the demosaiced image data being processed, to compensate for additional memory utilized by the demosaicing circuit when demosaicing certain types of image data. For example, when processing image data demosaiced from Quad Bayer image data, the front-end scaler circuit discards a portion of the chrominance component data for the received image data before performing chromatic suppression, compared to when processing image data demosaiced from Bayer image data.

Abstract: Techniques are disclosed for management of communication sessions of network traffic between client devices and the use of an up-to-date session state to enable seamless failovers between routers. One example technique may prepare each backup router to resume sessions of the active router in event of a failover and cause a redirection of the network traffic to complete the failover to a backup router. In a hot-switchover example, a network device known as a session controller synchronizes the session state information to backup router prior to failure and then, causes the network traffic to be redirected to backup router in response to the active router failure. In a warm-switchover example, the same session controller selects a backup router dynamically after detecting failure to active router, synchronizes session state information to backup router, and trigger routing updates, causing the network traffic to be redirected to the backup router.

Abstract: Embodiments relate to a multi-mode demosaicing circuit able to receive and demosaic image data in a different raw image formats, such as Bayer raw image format and Quad Bayer raw image format. The multi-mode demosaicing circuit comprises different circuitry for demosaicing different image formats that access a shared working memory. In addition, the multi-mode demosaicing circuit shares memory with a post-processing and scaling circuit configured to perform subsequent post-processing and/or scaling of the demosaiced image data, in which the operations of the post-processing and scaling circuit are modified based on the original raw image format of the demosaiced image data to use different amounts of the shared memory, to compensate for additional memory utilized by the multi-mode demosaicing circuit when demosaicing certain types of image data.

Abstract: Embodiments relate to a multi-mode demosaicing circuit able to receive and demosaic image data in a different raw image formats, such as Bayer raw image format and Quad Bayer raw image format. The multi-mode demosaicing circuit demosaics Quad Bayer image data by interpolating a green channel of the image data along each of a plurality of directions, generating a gradient of the image data along each of the plurality of directions, modifying the interpolated green channels based on respective gradients to generate full-resolution green channel image data, which is combined with red and blue image data to generate the demosaiced image data. Interpolation is performed for non-green pixels based on neighboring green pixels along a specified direction, modified by a residual value based upon values of one or more nearby same-color pixels and a correlation between values of the same-color pixels and neighboring green pixels.

Abstract: An approach for reducing disturbed errors in a flash memory device is disclosed. The approach includes collecting information associated with one or more; determining one or more frequently accessed data blocks from the one or more blocks based on the collected information; determining one or more neighboring blocks from the one or more blocks based on the collected information; determining if the one or more neighboring blocks exceeds a disturbance threshold; and in responsive to the one or more neighboring blocks has exceeded the disturbance threshold, re-align the one or more blocks.

Abstract: The present invention relates to the in vitro use of a group of cell-penetrating conjugate systems or a formulation incorporating the same for transporting e.g. therapeutically active agents, such as chemotherapeutics, insulin and immunosuppressants across the cellular membranes of mammalian cells, including human cells. The invention also relates to an in vitro methods of transporting physiologically active agents, such as therapeutics, across cell-membranes by using the cell-penetrating conjugate systems according to the present invention as well as the cell-penetrating conjugate systems for use in treatment of cancer, diabetes and for use in immunotherapy.

Abstract: In an approach to improve the field of multi-cloud environments by detecting data corruption between storage systems. Embodiments perform information tunneling on data transferring between a source storage system and a target storage system. Further, embodiments determine a checksum data of a data payload does not match an Internet Protocol (IP) packet extracted checksum and a blockchain based checksum and compare the checksum data at the target storage system with the IP packet extracted checksum and the blockchain based checksum to identify one or more checksum mismatches. Additionally, embodiments identify a corruption in a data payload based on the comparison between the checksum data at the target storage system and the IP packet extracted checksum and the blockchain based checksum, validate the corruption in the data payload, and update respective entities of identified corruption in the data payload.

Abstract: Aspects of the present invention disclose a method for dynamically adjusting a clock speed of a core of a multicore processor of user equipment. The method includes one or more processors generating a dedicated logical channel from a user device to a multidomain service orchestration layer of a fifth generation (5G) telecom network. The method further includes collecting information of the user device through the dedicated logical channel. The method further includes identifying situational insight of the user device based at least in part on the information of the user device. The method further includes identifying determining a workload forecast of one or more cores of the user device based at least in part on the situational insight. The method further includes identifying determining a recommended central processing unit (CPU) clock speed for a CPU core of the user device based at least in part on the workload forecast.

Abstract: The present disclosure relates to several techniques for reducing the occurrence of data collisions, which can occur when multiple devices simultaneously transmit data in the same (frequency) channel. For example, a turnaround time in which a device switches from a receive mode of operation to a transmit mode of operation may be reduced based on the device and another device indicating that they are capable of operating with a reduced turnaround. As another example, devices may use learning-based turnaround estimation to determine a turnaround time supported by other devices and utilize the determined turnaround time, for instance, instead of a longer turnaround time. As yet another example, multiple clear channel assessments may be performed before transmitting data. For instance, a first clear channel assessment may be performed during a backoff period, and a second clear channel assessment may be performed afterwards before data is transmitted.

Abstract: A process for the production of ultralow aromatic specialty distillate from different refinery streams particularly high sulfur streams. The process produces de-aromatized distillates with benzene content below 1 ppmw. Hydrocarbon feedstock having boiling temperature in the range of 90 and 350° C., preferably 140 and 320° C. The hydrocarbon feedstocks are obtained from any petroleum-refinery or bio-refinery or any other source producing hydrocarbon streams.

Abstract: A computer-implemented method includes: receiving, by a computing device, a query from a user device; generating, by the computing device and using a machine learning model, an answer to the query; determining, by the computing device, an urgency level of the query; generating, by the computing device, an explanation of the answer based on the determined urgency level; and presenting, by the computing device, the answer and the explanation to a user via the user device.

Abstract: The present disclosure relates to an ingestible device for sampling of the gastrointestinal tract. In particular, the device comprises an oral delivery capsule comprising a container having on the inside a hydrogel pre-polymer. The hydrogel prepolymer is capable of polymerising in situ, such as within the gastrointestinal tract, thereby collecting a sample of the gastrointestinal lumen by absorbing the lumen and its contents into the formed hydrogel.

Abstract: In an approach for in-band selective data snapshot using fifth generation (5G) radio link control (RLC) channels for edged cloud application backups, a processor creates one or more RLC channels to transfer network data packets in a telecommunication network using a network slicing technology. A processor allocates the one or more RLC channels to one or more applications in a user device for dedicated packet routing requirements. A processor maps the corresponding RLC channels and applications with corresponding identifiers, respectively. A processor monitors the one or more applications for backups. A processor, in response to receiving an application's backup request, creates a copy of a datastore associated with the application based on changes monitored in the one or more applications.

Abstract: A method for replicating interfaces in augmented reality for missing devices is provided. A processor identifies an expected position pattern of one or more wearable devices in conjunction with an augmented reality device based on an expected usage pattern of a user. A processor compares a current position pattern of the one or more wearables user to the expected position pattern. In response to the current position pattern not matching the expected position pattern, a processor retrieves an overlay corresponding to a use case indicated by the expected usage pattern. A processor renders the overlay in the augmented reality device, where the overlay is rendered in the expected position pattern for the one or more wearable devices relative to a view of the augmented reality device.