Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss.

Abstract: The present disclosure provides connection management techniques based on graph neural networks (GNN) and deep reinforcement learning (DRL) to optimize user association and load balancing. A graph structure of a communication network is considered for the GNN architecture and DRL is used to learn parameters of the GNN algorithm/model. Connection management is defined as a combinatorial graph optimization problem, and the DRL mechanism uses the underlying graph to learn weights of the GNN for an optimal user connections or associations. The connection management techniques can consider local network features to make better decisions to balance network traffic load while network throughput is also maximized. Implementations are provided based on edge computing frameworks include the Open RAN (O-RAN) architecture. Other embodiments may be described and/or claimed.

Abstract: Logic may quantize a primary neural network. Logic may generate, by a secondary neural network logic circuitry for a primary neural network logic circuitry, quantization parameters. The primary neural network logic circuitry may comprise a primary neural network with multiple layers trainable with an objective function. Each of the multiple layers of the primary neural network may comprise multiple tensors. The secondary neural network logic circuitry may comprise one or more secondary neural networks trainable with the objective function to output the quantization parameters to the tensors.

Abstract: Methods, apparatus, systems and articles of manufacture to determine topologies for networks are disclosed. An example a non-transitory computer readable medium comprises instructions that, when executed, cause a machine to at least: determine link capacities for a plurality of links between nodes of a network, determine a maximum number of children of the peer linked nodes, determine a maximum number of parents of the peer linked nodes, and utilize reinforcement learning to determine a subset of the plurality of links to be activated in the network based on the link capacities, the maximum number of children, and the maximum number of parents.

Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss.

Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss.

Abstract: Apparatuses and methods of manufacturing same, systems, and methods for object detection using a region-based deep learning model are described. In one aspect, a method is provided, in which a region proposal network (RPN) is used to identify regions of interest (RoI) in an image by assigning a confidence levels, the assigned confidence levels of the RoIs are used to boost the background score assigned by the downstream classifier to each RoI, and the background scores are used in a softmax function to calculate the final class probabilities for each object class.

Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss.

Abstract: A electronic system includes: a support chip configured to receive an input code stream; a circular Viterbi mechanism, coupled to the support chip, configured to: generate a final path metric for the input code stream, store intermediate path metrics at the repetition depth, generate a repetition path metric for the input code stream, and calculate a soft correlation metric based on the final path metric, the repetition path metric, and the intermediate path metrics.

Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss.

Abstract: Apparatuses and methods of manufacturing same, systems, and methods for object detection using a region-based deep learning model are described. In one aspect, a method is provided, in which a region proposal network (RPN) is used to identify regions of interest (RoI) in an image by assigning a confidence levels, the assigned confidence levels of the RoIs are used to boost the background score assigned by the downstream classifier to each RoI, and the background scores are used in a softmax function to calculate the final class probabilities for each object class.

Abstract: A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss.

Abstract: A electronic system includes: a support chip configured to receive an input code stream; a circular Viterbi mechanism, coupled to the support chip, configured to: generate a final path metric for the input code stream, store intermediate path metrics at the repetition depth, generate a repetition path metric for the input code stream, and calculate a soft correlation metric based on the final path metric, the repetition path metric, and the intermediate path metrics.

Abstract: Systems and methods for adaptive modulation and coding with frame size adjustment are described. In various implementations, these systems and methods may be applicable to Power Line Communications (PLC). For example, a method may include identifying a temporal region of a cyclostationary noise over which a frame is to be sent across a PLC network, the cyclostationary noise having a plurality of temporal regions, each of the plurality of temporal regions having a distinct spectral shape. The method may also include applying a given one of a plurality of Modulation and Coding Schemes (MCSs) to the frame to produce a modulated frame, wherein the given one of the plurality of MCSs is selected based, least in part, upon the spectral shape corresponding to the identified temporal region. The method may further include transmitting the modulated frame across the PLC network, at least in part, over the identified temporal region.

Abstract: Systems and methods for Carrier Sense Multiple Access (CSMA) and collision detection using a noise model are described. In various implementations, these systems and methods may be applicable to Power Line Communications (PLC). For example, a method may include receiving a signal via a communications channel in a PLC network, determining a feature of the signal, comparing the feature of the signal with a corresponding feature of a cyclostationary noise model, and taking a predetermined action based, at least in part, upon the comparison. In some implementations, taking the predetermined action may include determining whether to backoff or to transmit a packet over the communications channel. In other implementations, taking the predetermined action may include determining whether an error is due to a packet collision or due to a low quality of the communications channel.

Abstract: Systems and methods for adaptive modulation and coding with frame size adjustment are described. In various implementations, these systems and methods may be applicable to Power Line Communications (PLC). For example, a method may include identifying a temporal region of a cyclostationary noise over which a frame is to be sent across a PLC network, the cyclostationary noise having a plurality of temporal regions, each of the plurality of temporal regions having a distinct spectral shape. The method may also include applying a given one of a plurality of Modulation and Coding Schemes (MCSs) to the frame to produce a modulated frame, wherein the given one of the plurality of MCSs is selected based, least in part, upon the spectral shape corresponding to the identified temporal region. The method may further include transmitting the modulated frame across the PLC network, at least in part, over the identified temporal region.