Abstract: Approaches are presented for training and using scene graph generators for transfer learning. A scene graph generation technique can decompose a domain gap into individual types of discrepancies, such as may relate to appearance, label, and prediction discrepancies. These discrepancies can be reduced, at least in part, by aligning the corresponding latent and output distributions using one or more gradient reversal layers (GRLs). Label discrepancies can be addressed using self-pseudo-statistics collected from target data. Pseudo statistic-based self-learning and adversarial techniques can be used to manage these discrepancies without the need for costly supervision from a real-world dataset.

Abstract: Apparatuses, systems, and techniques to route data transfers between hardware devices. In at least one embodiment, a path over which to transfer data from a first hardware component of a computer system to a second hardware component of a computer system is determined based, at least in part, on one or more characteristics of different paths usable to transfer the data.

Abstract: A method dynamically selects one of a first sampling order and a second sampling order for a ray trace of pixels in a tile where the selection is based on a motion vector for the tile. The sampling order may be a bowtie pattern or an hourglass pattern.

Abstract: In various examples, image areas may be extracted from a batch of one or more images and may be scaled, in batch, to one or more template sizes. Where the image areas include search regions used for localization of objects, the scaled search regions may be loaded into Graphics Processing Unit (GPU) memory and processed in parallel for localization. Similarly, where image areas are used for filter updates, the scaled image areas may be loaded into GPU memory and processed in parallel for filter updates. The image areas may be batched from any number of images and/or from any number of single- and/or multi-object trackers. Further aspects of the disclosure provide approaches for associating locations using correlation response values, for learning correlation filters in object tracking based at least on focused windowing, and for learning correlation filters in object tracking based at least on occlusion maps.

Abstract: One embodiment of a method includes predicting one or more three-dimensional (3D) mesh representations based on a plurality of digital images, wherein the one or more 3D mesh representations are refined by minimizing at least one difference between the one or more 3D mesh representations and the plurality of digital images.

Abstract: The disclosure is directed to a process that can predict and prevent an audio artifact from occurring. The process can monitor the systems, processes, and execution threads on a larger system/device, such as a mobile or in-vehicle device. Using a learning algorithm, such as deep neural network (DNN), the information collected can generate a prediction of whether an audio artifact is likely to occur. The process can use a second learning algorithm, which also can be a DNN, to generate recommended system adjustments that can attempt to prevent the audio glitch from occurring. The recommendations can be for various systems and components on the device, such as changing the processing system frequency, the memory frequency, and the audio buffer size. After the audio artifact has been prevented, the system adjustments can be reversed fully or in steps to return the system to its state prior to the system adjustments.

Abstract: In various examples, systems and methods are disclosed relating to aligning images into frames of a first video using at least one first temporal attention layer of a neural network model. The first video has a first spatial resolution. A second video having a second spatial resolution is generated by up-sampling the first video using at least one second temporal attention layer of an up-sampler neural network model, wherein the second spatial resolution is higher than the first spatial resolution.

Abstract: Apparatuses, systems, and methods to move end connectors. In at least one embodiment, a linkage system to move an end connector between at least a first position and a second position is driven by an actuator in a first direction to drive movement of the end connector in a second direction, perpendicular to the first direction.

Abstract: In various examples, historical trajectory information of objects in an environment may be tracked by an ego-vehicle and encoded into a state feature. The encoded state features for each of the objects observed by the ego-vehicle may be used—e.g., by a bi-directional long short-term memory (LSTM) network—to encode a spatial feature. The encoded spatial feature and the encoded state feature for an object may be used to predict lateral and/or longitudinal maneuvers for the object, and the combination of this information may be used to determine future locations of the object. The future locations may be used by the ego-vehicle to determine a path through the environment, or may be used by a simulation system to control virtual objects—according to trajectories determined from the future locations—through a simulation environment.

Abstract: A portable computing device comprises: a base portion that includes a keyboard; and a display portion that is movably coupled to the base portion and includes: a heat sink with cooling fins; one or more heat-generating electronic devices that are thermally coupled to the heat sink; and at least one cooling fan configured to direct cooling air across the cooling fins.

Abstract: A computer-implemented method may comprise: receiving sensor data from a sensor of an autonomous vehicle; determining a presence of a lane closure object located on a lane element; determining a change of the lane closure object, selected from the presence of the lane closure object or absence of the lane closure object on the lane element; generating a change candidate based on the change in the lane closure object; obtaining a plurality of the change candidates during a time period or the autonomous vehicle being on a preceding lane element on the route; analyzing the plurality of change candidates for the change being the presence of the lane closure object or the absence of the lane closure object on the lane element; generating a final change candidate based on the change; and providing the final change candidate for updating a high definition map of the route having the lane element.

Abstract: Systems and methods for cooling a computer environment are disclosed. In at least one embodiment, a cooling assembly can be used to monitor and control fluid quality associated with one or more servers.

Abstract: Systems and methods for cooling in a datacenter are disclosed. In at least one embodiment, a thermal load bank (TLB) system to test a hybrid datacenter cooling system includes one or more thermal features to generate heat within a TLB system and includes one or more hybrid cooling features to provide air and liquid cooling responses to such heat generated by one or more thermal features.

Abstract: Apparatuses, systems, and techniques to perform non-maximum suppression (NMS) with a bit-reduced radix sort to remove redundant bounding boxes are described. In at least one embodiment, one or more circuits perform i) a bit-reduced radix sort operation to sort a list of confidence scores associated with a set of bounding boxes corresponding to one or more objects within one or more digital images and ii) a non-maximum suppression (NMS) operation on the sorted list to remove one or more redundant bounding boxes from the set.

Abstract: Approaches presented herein provide for unsupervised domain transfer learning. In particular, three neural networks can be trained together using at least labeled data from a first domain and unlabeled data from a second domain. Features of the data are extracted using a feature extraction network. A first classifier network uses these features to classify the data, while a second classifier network uses these features to determine the relevant domain. A combined loss function is used to optimize the networks, with a goal of the feature extraction network extracting features that the first classifier network is able to use to accurately classify the data, but prevent the second classifier from determining the domain for the image. Such optimization enables object classification to be performed with high accuracy for either domain, even though there may have been little to no labeled training data for the second domain.

Abstract: In various examples, systems and methods are disclosed relating to neural networks for realistic and controllable agent simulation using guided trajectories. The neural networks can be configured using training data including trajectories and other state data associated with subjects or agents and remote or neighboring subjects or agents, as well as context data representative of an environment in which the subjects are present. The trajectories can be determining using the neural networks and using various forms of guidance for controllability, such as for waypoint navigation, obstacle avoidance, and group movement.

Abstract: PUF cells utilizing a dual-interlocking scheme demonstrating improved noise immunity and stability across different V/T conditions and different uses over time in noisy environments. The PUF cell may be advantageously utilized in conjunction with error detection techniques that screen out unstable cells. A set of such PUF cells utilized to generate a device-specific bit pattern, for example a master key.

Abstract: Multi-ported memories that include write peripheral logic configured to operate in a first voltage domain, read peripheral logic configured to operate in a second voltage domain, and at least one bit cell array, wherein the write peripheral logic and the read peripheral logic are disposed on opposite sides of the bit cell array and voltage domain crossings between the first voltage domain and the second voltage domain are localized in bit cells of the at least one bit cell array.

Abstract: Mechanisms to exploit the inherent resiliency of deep learning inference workloads to improve the energy efficiency of computer processors such as graphics processing units with these workloads. The mechanisms provide energy-accuracy tradeoffs in the computation of deep learning inference calculations via energy-efficient floating point data path micro-architectures with integer accumulation, and enhanced mechanisms for per-vector scaled quantization (VS-Quant) of floating-point arguments.

Abstract: A system to generate a latent space model of a scene or video and apply this latent space and candidate sentences formed from digital audio to a vision-language matching model to enhance the accuracy of speech-to-text conversion. A latent space embedding of the scene is generated in which similar features are represented in the space closer to one another. An embedding for the digital audio is also generated. The vision-language matching model utilizes the latent space embedding to enhance the accuracy of transcribing/interpreting the embedding of the digital audio.