Abstract: In various examples, a machine learning model—such as a deep neural network (DNN)—may be trained to use image data and/or other sensor data as inputs to generate two-dimensional or three-dimensional trajectory points in world space, a vehicle orientation, and/or a vehicle state. For example, sensor data that represents orientation, steering information, and/or speed of a vehicle may be collected and used to automatically generate a trajectory for use as ground truth data for training the DNN. Once deployed, the trajectory points, the vehicle orientation, and/or the vehicle state may be used by a control component (e.g., a vehicle controller) for controlling the vehicle through a physical environment. For example, the control component may use these outputs of the DNN to determine a control profile (e.g., steering, decelerating, and/or accelerating) specific to the vehicle for controlling the vehicle through the physical environment.

Abstract: Systems and methods for an efficient and robust multiprocessor-coprocessor interface that may be used between a streaming multiprocessor and an acceleration coprocessor in a GPU are provided. According to an example implementation, in order to perform an acceleration of a particular operation using the coprocessor, the multiprocessor: issues a series of write instructions to write input data for the operation into coprocessor-accessible storage locations, issues an operation instruction to cause the coprocessor to execute the particular operation; and then issues a series of read instructions to read result data of the operation from coprocessor-accessible storage locations to multiprocessor-accessible storage locations.

Abstract: Machine learning is a process that learns a model from a given dataset, where the model can then be used to make a prediction about new data. In order to reduce the costs associated with collecting and labeling real world datasets for use in training the model, computer processes can synthetically generate datasets which simulate real world data. The present disclosure improves the effectiveness of such synthetic datasets for training machine learning models used in real world applications, in particular by generating a synthetic dataset that is specifically targeted to a specified downstream task (e.g. a particular computer vision task, a particular natural language processing task, etc.).

Abstract: A neural network structure is separated into an odd neural network including only the odd layers and an even neural network including only the even layers. In order to allow for parallel execution, for forward propagation a second input is generated from the original input, while for backward propagation a second error gradient is generated. Parallel execution may accelerate the forward and backward propagation operations without significant change in accuracy of the model. Additionally, restructuring a single neural network into two or more parallel neural networks may reduce the total time needed for training.

Abstract: Systems and methods related to transferring (e.g., large) files over a network are disclosed. In at least one embodiment, a client-server framework establishes a QUIC connection between a server application and a client application. Source files are processed by the server application to divide the source files into a number of chunks. Differential file transfer can be implemented between the client application and the server application by comparing metadata for chunks of the source file with metadata of local chunks of a destination file already stored in a local storage associated with the client application. Missing chunks can be requested from the server application and transferred to the client application using HTTP/3 messages.

Abstract: A glitch detection device includes an oscillator to generate multiple local clocks of multiple different phases and a sampling circuit to oversample, using the multiple local clocks, a system clock to generate multiple samples of the system clock. The device further includes digital logic that in turn includes a glitch detector to monitor a variation in pulse width of the system clock based on counting the multiple samples and to report a glitch in response to detecting a variation in the pulse width that exceeds a threshold value. The digital logic further includes a loop filter coupled between the glitch detector and the oscillator. The loop filter variably adjusts the oscillator based on a frequency of each of the multiple samples to control an output frequency of each of the multiple different phases of the oscillator.

Abstract: Systems and methods for detecting blockages in images are described. An example method may include receiving a plurality of images captured by a camera installed on an apparatus. The method may include identifying one or more candidate blocked regions in the plurality of images. Each of the candidate blocked regions may contain image data caused by blockages in the camera's field-of-view. The method may further include assigning scores to the one or more candidate blocked regions based on relationships among the one or more candidate blocked regions in the plurality of images. In response to a determination that one of the scores is above a predetermined blockage threshold, the method may include generating an alarm signal for the apparatus.

Abstract: A multi-level contrastive training strategy for training a neural network relies on image pairs (no other labels) to learn semantic correspondences at the image level and region or pixel level. The neural network is trained using contrasting image pairs including different objects and corresponding image pairs including different views of the same object. Conceptually, contrastive training pulls corresponding image pairs closer and pushes contrasting image pairs apart. An image-level contrastive loss is computed from the outputs (predictions) of the neural network and used to update parameters (weights) of the neural network via backpropagation. The neural network is also trained via pixel-level contrastive learning using only image pairs. Pixel-level contrastive learning receives an image pair, where each image includes an object in a particular category.

Abstract: A geometric approach may be used to detect objects on a road surface. A set of points within a region of interest between a first frame and a second frame are captured and tracked to determine a difference in location between the set of points in two frames. The first frame may be aligned with the second frame and the first pixel values of the first frame may be compared with the second pixel values of the second frame to generate a disparity image including third pixels. One or more subsets of the third pixels that have a value above a first threshold may be combined, and the third pixels may be scored and associated with disparity values for each pixel of the one or more subsets of the third pixels. A bounding shape may be generated based on the scoring.

Abstract: Technologies for low jitter and low power ring oscillators with multi-phase signal reassembly are described. A ring oscillator circuit includes a ring oscillator with a set of M delay stages, each stage outputs a phase signal, where M is a positive integer greater than one. The ring oscillator circuit includes a phase selector circuit coupled to the ring oscillator. The phase selector circuit can receive M phase signals from the ring oscillator and generate N phase signals based on the M phase signals, where N is a positive integer less than M.

Abstract: Methods and apparatus are described for detecting anomalies in a clock signal. Example methods include sensing a clock signal that exhibits alternating phases during normal operation; responsive to sensing the start of a first phase, generating a pulse; and if the pulse terminates before sensing the end of the first phase, asserting a clock stopped detection signal. Example clock anomaly detection apparatus includes a clock signal input for coupling to a clock signal that, during normal operation, oscillates between first and second clock states. An anomaly detection output is asserted if the clock signal remains in the first clock state longer than a first phase expected duration or remains in the second clock state longer than a second phase expected duration.

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: Disclosed approaches provide for interactions of secondary rays of light transport paths in a virtual environment to share lighting contributions when determining lighting conditions for a light transport path. Interactions may be shared based on similarities in characteristics (e.g., hit locations), which may define a region in which interactions may share lighting condition data. The region may correspond to a texel of a texture map and lighting contribution data for interactions may be accumulated to the texel spatially and/or temporally, then used to compute composite lighting contribution data that estimates radiance at an interaction. Approaches are also provided for reprojecting lighting contributions of interactions to pixels to share lighting contribution data from secondary bounces of light transport paths while avoiding potential over blurring.

Abstract: In various examples, a deep neural network(s) (e.g., a convolutional neural network) may be trained to detect moving and stationary obstacles from RADAR data of a three dimensional (3D) space. In some embodiments, ground truth training data for the neural network(s) may be generated from LIDAR data. More specifically, a scene may be observed with RADAR and LIDAR sensors to collect RADAR data and LIDAR data for a particular time slice. The RADAR data may be used for input training data, and the LIDAR data associated with the same or closest time slice as the RADAR data may be annotated with ground truth labels identifying objects to be detected. The LIDAR labels may be propagated to the RADAR data, and LIDAR labels containing less than some threshold number of RADAR detections may be omitted. The (remaining) LIDAR labels may be used to generate ground truth data.

Abstract: Systems and methods for cooling a data center are disclosed. In at least one embodiment, a radiator is associated with one or more racks of a datacenter and has a first portion to function as an air-to-liquid heat exchanger having a primary cooling loop to absorb first heat away from the one or more racks and has a second portion to function as a liquid-to-liquid heat exchanger to enable at least one secondary cooling loop to exchange second heat with the primary cooling loop.

Abstract: A framework for offline learning from a set of diverse and suboptimal demonstrations operates by selectively imitating local sequences from the dataset. At least one embodiment recovers performant policies from large manipulation datasets by decomposing the problem into a goal-conditioned imitation and a high-level goal selection mechanism.

Abstract: Approaches in accordance with various embodiments provide for fluid simulation with substantially reduced time and memory requirements with respect to conventional approaches. In particular, various embodiments can perform time and energy efficient, large scale fluid simulation on processing hardware using a method that does not solve for the Navier-Stokes equations to enforce incompressibility. Instead, various embodiments generate a density tensor and rigid body map tensor for a large number of particles contained in a sub-domain. Collectively, the density tensor and rigid body map may represent input channels of a network with three spatial-dimensions. The network may apply a series of operations to the input channels to predict an updated position and updated velocity for each particle at the end of a frame. Such approaches can handle tens of millions of particles within a virtually unbounded simulation domain, as compared to classical approaches that solve for the Navier-Stokes equations.

Abstract: Systems and methods for cooling a datacenter are disclosed. In at least one embodiment, fins are provided within a cold plate and are adjustable to control an amount of surface area of the fins to be exposed to a fluid and to be cooled by the fluid based, at least in part, upon a temperature associated with the fluid or with at least one computing device.

Abstract: A neural network system leverages dual attention, specifically both spatial attention and channel attention, to jointly estimate heart rate and respiratory rate of a subject by processing images of the subject. A motion neural network receives images of the subject and estimates heart and breath rates of the subject using both spatial and channel domain attention masks to focus processing on particular feature data. An appearance neural network computes a spatial attention mask from the images of the subject and may indicate that features associated with the subject's face (as opposed to the subject's hair or shoulders) to accurately estimate the heart and/or breath rate. Channel-wise domain attention is learned during training and recalibrates channel-wise feature responses to select the most informative features for processing. The channel attention mask is learned during training and can be used for different subjects during deployment.

Abstract: A system includes a link having one or more lanes associated with transmitting data and one or more lanes associated with transmitting a clock signal. The system includes a device coupled with the link, the device to receive a signal via the one or more lanes associated with transmitting the clock signal and determine a number of pulses associated with the signal over a period. The device is further to determine the number of pulses associated with the signal fail to satisfy a predetermined condition relating to a specified number of pulses for the period and initiate a power-down sequence in response to determining the number of pulses that fail to satisfy the predetermined condition relating to the specified number of pulses for the period.