Abstract: This disclosure describes methods to process timing information of flows in a network. One or more processors determine a latency associated with each of one or more packets of a flow passing through a device. The one or more processors determine that the latency is greater than a baseline latency, and the one or more processors provide a message indicating at least the flow and that the latency is greater than the baseline latency.

Abstract: A generator network of a variational autoencoder can be trained to approximate a simulator and generate a first result. The simulator is associated with input data, based on which the simulator outputs output data. A training data set for the generator network can include the simulator's input data and output data. Based on the simulator's output data and the first result of the generator network, an inference network of the variational autoencoder can be trained to generate a second result. The second result of the trained inference network inverts the first result of the generator and approximates the simulator's input data. The trained inference network can function as an inverted simulator.

Abstract: Methods and server systems for detecting fraudulent transactions are described herein. Method performed by server system includes accessing base graph including plurality of nodes further including plurality of labeled nodes and unlabeled nodes. Method includes assigning via Graph Neural Network (GNN) model, fraudulent label or non-fraudulent label to each unlabeled node based on the base graph. This assigning process includes generating plurality of sub-graphs based on splitting the base graph and filtering these sub-graphs via Siamese Neural Network model based on pre-defined threshold values. Then, the GNN model generates plurality of sets of embeddings based on plurality of filtered sub-graphs. Further, aggregated node embedding is generated for each node and then, final node representation for each node is generated via dense layer of GNN model. Then, fraudulent label or the non-fraudulent label is assigned to each unlabeled node of plurality of unlabeled nodes based on final node representation.

Abstract: Techniques are provided for computing with private healthcare data. The techniques include a method comprising constructing an isolated memory partition that forms a secure enclave and pre-provisioning software within the secure enclave. The pre-provisioned software is configured to receive at least one of input data or the instructions for the one or more application computing processes in an encrypted form; decrypt the at least one of input data or instructions using one or more cryptographic keys; execute the one or more application computing processes based on the decrypted at least one of input data or instructions to generate output data; generate a proof of execution that indicates that the one or more application computing processes operated on the received input data; encrypt the output data using the one or more cryptographic keys; and provide external access to the encrypted output data and the proof of execution.

Abstract: Methods and systems for securing an application programming interface (API) are presented. The method comprises: receiving API workflow data associated with an API testing tool and generating a scan configuration file using the API workflow data; crawling the collection of API requests by identifying and retrieving a link associated with the collection of API requests; and crawling the link to generate a crawled link response. The method also includes executing one or more vulnerability tests on the crawled link response including applying at least one passive detection rule to the crawled link response and fuzzing the link. The fuzzed link may be transmitted in a request to an application server following which scan data indicative of at least one vulnerability associated with a response from the application server may be generated. The scan data may be used to generate a vulnerability report.

Abstract: A UE may determine use of a sensor or an interface of the UE, and in response, may adjust at least one of a beam selection, a transmission power, an operation mode, or a performance of carrier aggregation for the UE. A base station may receive an indication from the UE that a carrier for which the UE is configured has an effect on performance of a sensor or an interface of the UE. The base station may adjust a configuration of the carrier or adjusting uplink grants for the carrier in response to receiving the indication from the UE.

Abstract: A method for managing designated authority status in a cache line includes identifying an initial designated authority (DA) member cache for a cache line, transferring DA status from the initial DA member cache to a new DA member cache, determining whether the new DA member cache is active, indicating a final state of the initial DA cache responsive to determining that the new DA member cache is active, and overriding a DA state in a cache control structure in a directory. A method for managing cache accesses during a designated authority transfer includes receiving a designated authority (DA) status transfer request, receiving an indication that a first cache will invalidate its copy of the cache line, allowing a second cache to assume DA status for the cache line, and denying access to the first cache's copy of the cache line until invalidation by the first cache is complete.

Abstract: Techniques for implementing a dynamic intelligent log analysis tool are disclosed. In some embodiments, a computer system performs operations comprising: obtaining a log file comprising a plurality of log entries, each log entry comprising an error message; identifying a set of unique words from the error messages; for each error message, computing a term-frequency vector based on a frequency of occurrence for each unique word of the set of unique words in the error message; for each error message, computing a similarity measure between the term-frequency vectors of the error message and every other error message of the log entries; for each error message, computing a score based on a sum of the similarity measures; and displaying an indication of one or more of the error messages on a computing device based on the scores for the one or more of the error messages.

Abstract: A method comprises receiving an input specifying one or more rules in connection with archiving one or more of a plurality of files from a source storage location to a target storage location. The one or more rules specify one or more constraints for retention of the one or more of the plurality of the files. In the method, the one or more of the plurality of files are retrieved from the source storage location for migration to the target storage location. A request is sent to the target storage location that the target storage location invoke the one or more rules to retain the one or more of the plurality of files in the target storage location. The target storage location comprises a cloud storage platform.

Abstract: A system and method for detecting rider impairment based on image or audio input is implemented in a rental fleet of lightweight vehicles. The system comprises a mobile device, a backend server, and one or more lightweight vehicles. Access to fleet vehicles is controlled by the mobile application based on results of data collected about the prospective driver.

Abstract: The present disclosure relates to a system and method for automatic vehicle recognition, based on a smart device. The system mainly includes an image capturing device integrated in the smart device, a data storage to store the images captured by the image capturing device and known identity aspects related to a vehicle allotted to a user, and a License plate processing and matching (LPPM) component to perform recognition process. LPPM component includes an identity aspect detector to detect a portion representing the identity aspect, an image processor to enhance the portion and perform Optical Characters Recognition (OCR) to extract character strings from the portion. The character string is compared against a character string of the known identity aspect to verify the vehicle.

Abstract: A method includes obtaining video content and associated substantially mono audio content. The method also includes determining at least one of a position or a motion trajectory of each of one or more objects detected in the video content and classifying each of the one or more objects into one of multiple object classes. The method further includes separating audio streams within the audio content based on the video content. Each of the audio streams is associated with one of multiple audio sources. The method also includes classifying each of the audio sources into one of the object classes. In addition, the method includes, for each audio source classified into the same object class as one of the one or more objects, distributing the audio stream associated with that audio source into multiple audio channels based on at least one of the position or the motion trajectory of that object.

Abstract: An image classification system and method are used to determine the status of equipment completeness of the rental two-wheeled vehicles, such as electric scooters. The system and method use deep learning models to analyze and classify ambiguous states of the rental vehicle when the user finishes the ride. These states are likely to be encountered by rental vehicles, to protect helmets, trunks or other equipment from loss or damage.

Abstract: An XR helmet collects user data and environmental data using onboard sensors. The collected data is used to train a model for predicting safe rides in a rental fleet of lightweight vehicles. The XR helmet is configured to selectively take control of the lightweight vehicle when driver biofeedback or environmental data are predictive of unsafe driving conditions. Access and control of fleet vehicles is controlled by a predictive machine learning model trained by data indicative of safe and unsafe driving conditions.

Abstract: A burst mode time division multiple access (TDMA) data processing system can determine burst time plans (BTPs). The burst mode TDMA data processing system can receive burst mode time division multiple access (TDMA) data. The burst mode TDMA data can be distributed among processors based on the BTPs. The distributed burst mode TDMA data can be demodulated and decoded the using the processors and the demodulated and decoded burst mode TDMA data can be output.

Abstract: A system and method for rider profiling for lightweight vehicles is disclosed herein. The system comprises a lightweight vehicle actuating unit. Load-cells are configured on a deck of the lightweight vehicle for sensing and measuring load acting thereupon for generating a measurement signal. A pressure-pattern collection unit receives the measurement signal to identify and process a load pattern acting on the deck to generate a pressure-pattern signal. A learning model receives the pressure-pattern signal for processing to obtain information associated with a pose of the rider. A riding control is coupled to a riding control database and configured to unit receive the information associated with the pose from the learning model for determining at least one instance of rule violation and computing a decision based on the at least one instance of rule violation.

Abstract: A smart system for detecting the presence of a helmet in a vehicle is disclosed herein. The system comprises a first transceiver configured within a trunk of the vehicle. The first transceiver is configured to transmit a first detection signal on detecting the opening of the trunk of the vehicle. At least one second transceiver corresponds to at least one helmet provided on the body of the helmet. The second transceiver is configured to respond to the first detection signal with a response signal, wherein the response signal includes a unique identity marker corresponding to the helmet. A transmission device is configured to detect opening and closing of trunk to activate the first transceiver, subsequent to which the first detection signal is sent, and the transmission device is further configured to transmit information associated with the response signal to a base controller.

Abstract: The present invention discloses a Double Metal Cyanide (DMC) catalyst(s) useful for the production of polyether polyols (PEPO) and a less energy intensive room temperature method for the synthesis thereof. The catalyst(s) comprises of a DMC complex, an organic complexing agent, i.e., ethylenediaminetetraacetic acid (EDTA) and other co-complexing organic agents, e.g., t-BuOH, PEPO of composition ranging from about 1 to 10 wt %, wherein the average molecular weight of PEPO used ranged from 200 to 1000. A method of preparing a series of DMC catalyst(s) at room temperature with varying compositional ratios of the complexing and co-complexing agents targeting a wide range of PEPO of varying kinematic viscosity range is also disclosed. These DMC catalyst(s) are amorphous, highly active, and easily separable from product PEPO with recyclability/recoverability, making the product PEPO better industrially applicable and DMC catalyst more cost-effective.

Abstract: A 3D printing and assembly system includes a 3D printer having a build volume; a robotic arm configured to access both within the build volume and outside of the printer. The printing and assembly system and a 3D computer hardware system are connected to both the printer and the robotic arm. An assistive object outside of build volume and accessible by robotic arm is identified. A 3D object assembly to be generated by the printer is identified. The assistive object and the object assembly is real-time analyzed, using the computer hardware system, to generate interdependent sequential instructions for the printer and the robotic arm. The already-generated object is positioned within the build volume using the robotic arm with the sequential instructions for the robotic arm. The object assembly is 3D printed by 3D printing around the already-generated object using the sequential instructions for the 3D printer.

Abstract: An embodiment for dynamically displaying images based on textual content is provided. The embodiment may include receiving real-time textual content from a sender and a recipient. The embodiment may also include deriving a topic of the real-time textual content and a context associated with the real-time textual content. The embodiment may further include identifying a medium in which the real-time textual content is entered. The embodiment may also include predicting a digital image to be displayed to the sender. The embodiment may further include in response to determining the digital image is available in a digital library, selecting the digital image from the digital library and displaying the selected digital image to the sender. The embodiment may also include validating the selected digital image and displaying the validated selected digital image to the recipient.