Abstract: A method for computer memory access includes, during execution of a machine learning model, receiving an input vector for multiplication with a matrix of network weight values. Each network weight value of the matrix of network weight values is stored in computer memory using a stored quantity(S) of bits. For a network weight value of the matrix of network weight values, a representation quantity (R) of bits is determined to be used for representing the network weight value during multiplication with a corresponding vector value of the input vector, based at least in part on a magnitude of the corresponding vector value. The R bits of the network weight value are retrieved from the computer memory for multiplication with the corresponding vector value.

Abstract: An operational mode can be designated for a computing device, such as an integrated circuit. In particular, an input signal may be used to determine an operational mode, which may lead to bypassing or otherwise restricting one or more boot operations to permit use of the computing device. Such an approach provides improved flexibility to permit use of the computing device for testing or debugging while maintaining security features used during other operational modes.

Abstract: Certain aspects of the present disclosure provide techniques for interference management reference signal (RS) transmission. An example method, performed at a first network entity, generally includes applying a frequency domain function to a first copy of a first reference signal (RS) sequence to obtain a modified first copy of the first RS sequence, and outputting, to a second network entity, frequency domain samples of the modified first copy of the first RS sequence and frequency domain samples of a second copy of the first RS sequence.

Abstract: A property interactive system includes the seller to display seller added listings. The property interactive system further includes a review property module. The property interactive system further includes a buyer the ability for immediate purchase of a listing using an on-chain title, the seller inspects pending buyer offers through to either accept the pending offer or deny the pending offer.

Abstract: A memory where video content is stored for access by processing components in a display pipeline is divided into different categories or groupings, each different category or grouping corresponding to a different security level. Access, by the processing components in the display pipeline, to the video content stored in the different categories or groupings is restricted in different ways so that access to video content stored in the highest security categories or groupings is more restricted and more secure than access to the video content stored in a less secure categories or groupings. Video content is received and a security level corresponding to video content is identified. The video content is written into a memory category or grouping, of the plurality of different categories or groupings corresponding to a plurality of different security levels, based upon the security level corresponding to the video content.

Abstract: Methods for integrating copper-graphene laminate (CGL) in a multilayer PCB fabrication process and the resulting lamination stacks are disclosed. The methods include providing a core and applying a first graphene layer to the surface of the core. The methods further include applying a metal layer to the first graphene layer and applying a second graphene layer to the metal layer. Further, the methods include applying a photoresist layer to the second graphene layer and applying a protective layer to the photoresist layer. In some embodiments, the methods include applying a metallic plating to lamination stack. The methods further include drilling through the protective layer and at least one of a photoresist layer, the second graphene layer, the metal layer, the first graphene layer, and/or the core.

Abstract: A rule training system and methods are provided that are configured to automatically generate machine learning (ML) rules for intelligent decision-making by a policy manager platform. The system includes a processor and a computer readable medium operably coupled thereto, the computer readable medium comprising a plurality of instructions stored in association therewith that are accessible to, and executable by, the processor, to perform rule training operations which include accessing rule training data, iteratively generating a plurality of decision rules based on the rule training data and a plurality of ML model training techniques, testing each of the plurality of decision rules, filtering the plurality of decision rules by corresponding performances, selecting a set of the plurality of decision rules based on alert metrics, evaluating the set of the plurality of decision rules, and generating a decision ruleset for the ML task.

Abstract: Methods of forming a graphene integrated core for making a printed circuit board (PCB) having enhanced thermal management properties are disclosed. The methods include providing a core body having a core body length and applying a graphene multi-layer to the core body to form a laminated stack, where the graphene multi-layer has a graphene multi-layer length that is shorter than the core body length. At least one conductive layer may be applied to the laminated stack. The graphene multi-layer may be disposed within the graphene integrated core such that the graphene multi-layer is electrically insulated from the at least one conductive layer. Corresponding graphene integrated cores having a graphene multi-layer that is disposed within the graphene integrated core such that the graphene multi-layer is electrically insulated from the at least one conductive layer are also described.

Abstract: A method of generating an aggregate saliency map using a convolutional neural network. Convolutional activation maps of the convolutional neural network model are received into a saliency map generator, the convolutional activation maps being generated by the neural network model while computing the one or more prediction scores based on unlabeled input data. Each convolutional activation map corresponds to one of the multiple encoding layers. The saliency map generator generates a layer-dependent saliency map for each encoding layer of the unlabeled input data, each layer-dependent saliency map being based on a summation of element-wise products of the convolutional activation maps and their corresponding gradients. The layer-dependent saliency maps are combined into the aggregate saliency map indicating the relative contributions of individual components of the unlabeled input data to the one or more prediction scores computed by the convolutional neural network model on the unlabeled input data.

Abstract: The system can be configured to execute a subset of “hooks” or “proxies,” to manage IS, DOM, or API execution. The subset of hooks or proxies are associated with known data access operations, known functions, etc., that target system accesses representing vulnerability or data accesses for sensitive information. For example, the subset leverage options available in conventional browsers to secure control as the system cannot recompile the end-user's browser in a real-time site scenario, nor force an extension installation. In such on-line environments, the system is configured to employ IS level proxies/hooks, that can be delivered, for example, by a IS tag, inline code, or a server-side templating solution. Based on execution of the real-time permission and access control, for example, via the IS level proxies/hooks, various embodiments are configured to provide real time alerts and restrictions on third party IS code, DOM requests, and APIs.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a reconfigurable intelligent surface (RIS) may select a weight vector, from a plurality of stored weight vectors, for enabling a spatially multiplexed communication, between a transmitter and a receiver, that includes two or more layers for each polarization of a plurality of polarizations for the spatially multiplexed communication. The RIS may receive the spatially multiplexed communication from the transmitter. The RIS may transmit the spatially multiplexed communication to the receiver in accordance with the weight vector. Numerous other aspects are described.

Abstract: In one embodiment, a method includes transmitting a plurality of beamformed reference signals using a corresponding plurality of beamforming precoders; receiving from each of the plurality of UEs a signal quality metric for at least a subset of the plurality of beamformed reference signals; determining a beam-specific cross-correlation for pairs of the plurality of UEs using the signal quality metrics; determining one or more scheduling groups of UEs from the plurality of UEs based on the beam-specific cross-correlation for pairs of the plurality of UEs; and allocating a set of time-frequency resource elements to two or more UEs selected from one of the determined scheduling groups.

Abstract: A computer can receive, by a software program executing on the processing device, a specification file drafted by a developer of a software application, the specification file having a declarative format and indicating a series of data retrieval tasks to be performed as part of a debugging process for debugging a problem with the software application. The computer can determine operations based on the series of data retrieval tasks indicated by the specification file, the operations being executable to perform the tasks. The computer can execute the one or more operations to perform the data retrieval tasks. By performing the data retrieval tasks, the computer can retrieve debugging information associated with one or more containers associated with the software application. The computer can transmit the retrieved debugging information to a debugging entity for use in debugging the problem with the software application.

Abstract: Examples are disclosed for methods and mechanisms to address Base Transceiver Stations (BTSes) attempting to hijack Mobile Stations (MSes) in wireless communication networks. In some example methods, a record having values for a set of features associated with an MS and a BTS of a wireless communication net may be processed. A score for the record may be calculated, for example based on a machine-learning model pertaining to the set of features, and a determination may be made as to whether the record demonstrates an anomalous condition based upon the score. The set of features may include at least one feature based on a measured power of the BTS.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a network node may obtain position information associated with a first subset of network node positions, of a set of network node positions, in connection with a map associated with a wireless coverage area. The network node may obtain map information associated with the map and may determine, based on a machine learning component, a first plurality of predicted pathloss values associated with a second subset of network node positions of the set of network node positions. Numerous other aspects are described.

Abstract: Apparatus and methods are described for use with a venous system of a subject that includes one or more tributary vessels that flow into a vein at junctions. A blood-pump-catheter (70) includes a material (36) configured to be placed at a downstream location that is downstream of the junctions and to at least partially occlude blood flow through the vein. The material defines a blood-outlet opening (31). A blood pump (24) pumps blood from a region of the vein that is adjacent to the junctions, and through the blood-outlet opening (31). A blood-flow pathway (60) directs upstream venous blood flow from an upstream location that is upstream of the junctions, to downstream of the material (36), without the upstream venous blood flow being pumped by the pump (24). Other applications are also described.

Abstract: The invention provides one or more consortia of networks that identify and share information about users and/or user devices interacting with the consortia. User devices may be identified, at least in part, by tag-based computer information. Computers and other devices accessing the Web carry device tags with date and time information describing when they were issued by a security tag server. A server time stamp may be inserted into time based computer tags such as a cookies indicating when they were created. Such time stamp information can be encrypted and analyzed during future attempts to access a secure network such as a customer attempting to log into an online banking account. When the time stamp information from the tag is compared to other selected information about the user, device and/or account, including but not limited to last account log-in date/time or account creation date, the invention may be used to detect suspicious activity.

Abstract: A pulse generation circuit in a quantum controller operates synchronously with a pulse computation circuit. The pulse generation circuit generates a pulse associated with a quantum element operation. The pulse computation circuit is able to determine characteristics of a signal that is based on the pulse. These characteristics are used by the pulse generation circuit to modify the pulse.

Abstract: According to some embodiments. a management system is configured to host a secure execution environment. Browser scripts and API requests can be evaluated in the secure execution environment to validate their functions in terms of security and/or data privacy. In further embodiments. the system is configured to validate operation via a proxy layer or hooking all of the function controls so that whenever is requested (e.g., outside resources or data). the system is configured to track what is requested and what operations are done. For example. once proper function has been validated in terms of data privacy and/or security. the candidate script. API. etc., can be signed as valid (e.g., via a secure hash). The secure hash can be used in subsequent operation to ensure that the script. API, etc. matches a known valid version and function.

Abstract: Systems and methods are described herein for providing power for enabling electroluminescence imaging of photovoltaic panels. The system may comprise a diode in a power converter, where the diode may restrict reverse current flow to the photovoltaic panel. The system may comprise a power device configured to be coupled to a photovoltaic panel. The power device may comprise an auxiliary power circuit which may provide power to the power device from the photovoltaic panel or form a power source connected to a power system controller. The power device may control a switch to provide a current path for reverse current to flow to the photovoltaic panel. An imager may capture an image of the panel.