Abstract: A processor – such as within a system-on-a-chip (SoC) – can keep memory work for different machine safety functions from interfering with each other. When the processor receives memory requests from different functions, it checks an identifier for each request that indicates the request’s associated safety level. Based on that identifier, the processor sends each request to a separate memory manager assigned to that safety level. By using different memory managers for different safety levels, the processor can handle memory operations for one safety function without disrupting memory operations for another safety function.

Abstract: In examples, the number of rays used to sample lighting conditions of a light source in a virtual environment with respect to particular locations in the virtual environment may be adapted to scene conditions. An additional ray(s) may be used for locations that tend to be associated with visual artifacts in rendered images. A determination may be made on whether to cast an additional ray(s) to a light source for a location and/or a quantity of rays to cast. To make the determination variables such as visibilities and/or hit distances of ray-traced samples of the light source may be analyzed for related locations in the virtual environment, such as those in a region around the location (e.g., within an N-by-N kernel centered at the location). Factors may include variability in visibilities and/or hit distances, differences between visibilities and/or hit distances relative to the location, and magnitudes of hit distances.

Abstract: Aspects of this technical solution can increase speed of processing in low-latency application areas, while maintaining integrity of image feature recognition at those higher speeds. For example, in image-processing environments associated with autonomous navigation (e.g., driving), a large volume of image data is to be rapidly and accurately processed to maintain reliable and up-to-date models of a physical environment. For example, embodiments in accordance with this disclosure can provide high-speed and accurate image feature recognition of input frame data beyond the capability of CPU processing or general GPU processing to achieve.

Abstract: A processor can update data by using many small processing elements that work together. Each processing element can receive data for one data element, use nearby data from one or more other data elements, and follow instruction data telling it what operation to perform. The processing element then calculates an updated value and outputs data for that updated value. This allows an array of processing elements, such as in an accelerator, to efficiently perform operations that depend on both a data element and surrounding data elements, such as image filtering or other complex data-processing tasks.

Abstract: A secure electronic component assembly is described herein for ensuring the physical integrity of an integrated circuit (IC). The secure electronic component assembly may comprise a printed circuit board (PCB), an integrated circuit (IC) mounted on the PCB, and an underfill material disposed between the IC and the PCB. The underfill material comprises a detection agent that is configured to change a state of the IC in response to exposure to an external environment, wherein the change in the state of the IC is indicative of a tamper condition.

Abstract: In various examples, ultrasonic data augmentations for autonomous and/or semi-autonomous systems and applications are described herein. Systems and methods described herein may use sensor data generated using one or more ultrasonic sensors to generate augmented input data for training one or more machine learning models to generate one or more representations (e.g., one or more maps) of an environment. As described herein, the sensor data may be augmented using one or more techniques such that the augmented input data corresponds to various driving environments (e.g., different driving surfaces), various poses on machines (e.g., different locations and/or orientations), and/or includes additional information associated with the ultrasonic sensor(s) and/or the sensor data.

Abstract: In various examples, sets of testing data may be selected and applied to an MLM such that differences in performance of the MLM in the testing between the sets indicates and may be used to determine whether and/or an extent by which the MLM is trained to rely on artifacts. Training data for the MLM may be generated using a first value of a parameter that defines a value of a characteristic of the training data. For testing, first testing data may be selected that corresponds to a second value of the parameter that shifts the value in a first direction and second testing data may be selected that corresponds to a third value of the parameter that shifts the value in a second direction (e.g., opposite the first direction). Various possible actions may be taken based on results of analyzing the differences in performance.

Abstract: Apparatuses, systems, and techniques to generate one or more hyperparameters to be used by one or more neural networks. In at least one embodiment, one or more hyperparameters to be used by one or more neural networks are generated based on, for example, a neural architecture search (NAS) algorithm.

Abstract: Systems and methods are disclosed relating to receiving at least one of a text prompt or a kinematic constraint, generating, by a motion model including a first model and a second model, human motion data of a human character by applying a random noise and the at least one of the text prompt or the kinematic constraint into the motion model. Generating the human motion data includes, for each iteration of diffusion determining, using the first model, global root motion by applying noisy global root motion and noisy local joint motion as inputs into the first model and determining, using the second model, local joint motion by applying the noisy local joint motion and local root motion as inputs into the second model. The local root motion is determined based on the global root motion. The human motion data includes the local joint motion and the global root motion.

Abstract: In various examples, a representative subset of data points are queried or selected using integer programming to minimize the Wasserstein distance between the selected data points and the data set from which they were selected. A Generalized Benders Decomposition (GBD) may be used to decompose and iteratively solve the minimization problem, providing a globally optimal solution (an identified subset of data points that match the distribution of their data set) within a threshold tolerance. Data selection may be accelerated by applying one or more constraints while iterating, such as optimality cuts that leverage properties of the Wasserstein distance and/or pruning constraints that reduce the search space of candidate data points. In an active learning implementation, a representative subset of unlabeled data points may be selected using GBD, labeled, and used to train machine learning model(s) over one or more cycles of active learning.

Abstract: Systems and methods for sharing security capabilities of network devices are disclosed. A system for a first network device includes a memory. The system also includes one or more processors, coupled to the memory, to receive, at the first network device, security capabilities of a second network device of a network. The network includes the first network device and the second network device. The one or more processors are further to modify a routing table of the first network device based on the security capabilities of the second network device and transmit a data packet based on the modified routing table.

Abstract: Approaches of the disclosure are directed towards the intelligent management of notifications. Notifications can be intelligently managed across digital devices by contextualizing and prioritizing notifications to, for example, match a current state or situation of a user. Incoming notifications may be contextualized by analyzing their content and sources, such as by using large language models. Such an approach may further take into account the user's current status, including factors such as location, activity, and personal preferences, as may be obtained from various sources or learned over time. Preferences or appropriate delivery methods can be learned by observing and/or analyzing user interactions associated with previously-presented notifications and adjusting the delivery methods for subsequent notifications, which may involve suppressing the notification, presenting immediately, changing an alert type, or altering content for presentation.

Abstract: Systems and methods for improved media stream processing. In at least one embodiment, a first media stream is assigned a hardware processing engine and a second media stream is assigned to a software processing engine based on a performance state of an application server, one or more parameters of the first media stream, and one or more parameters of the second media stream.

Abstract: In various examples, behavior-based mission task management for mobile autonomous machine systems and applications are provided. A mission controller may generate a mission behavior model logic framework for a mission that accounts for the actions of a set of multiple task agents that play a role in completing the mission. The mission controller may assemble a framework starting from a mission template that defines a baseline task sequence, correlate tasks defined by the baseline task sequence with pre-defined behavior models from a task library, and customize those behavior models based on mission task customization parameters. The mission controller may provide the mission behavior model logic framework to a mission dispatch function. Individual autonomous mobile task agents may then proceed to execute their assigned portions of local task sequences in accordance with the customized behavior models distributed to them by the mission dispatch function.

Abstract: Apparatuses, systems, and techniques to identify a location of one or more portions of incomplete graph code. In at least one embodiment, a location of one or more portions of incomplete graph code is identified based on, for example, CUDA or other parallel computing platform code.

Abstract: Systems and methods are disclosed that relate to synthesizing high-resolution 3D geometry and strictly view-consistent images that maintain image quality without relying on post-processing super resolution. For instance, embodiments of the present disclosure describe techniques, systems, and/or methods to scale neural volume rendering to the much higher resolution of native 2D images, thereby resolving fine-grained 3D geometry with unprecedented detail. Embodiments of the present disclosure employ learning-based samplers for accelerating neural rendering for 3D GAN training using up to five times fewer depth samples, which enables embodiments of the present disclosure to explicitly “render every pixel” of the full-resolution image during training and inference without post-processing super-resolution in 2D.

Abstract: Apparatuses, systems, and techniques to perform registration among images. In at least one embodiment, one or more neural networks are trained to indicate registration of features in common among at least two images by generating a first correspondence by simulating a registration process of registering an image and applying the at least two images and the first correspondence to a neural network to derive a second correspondence of the features in common among the at least two images.

Abstract: Embodiments of the present disclosure may include a method and system for performing one or more operations based on a visibility confidence model. In some embodiments, the method may include generating a visibility confidence model which may indicate a level of confidence in sensor data that may correspond to individual sub-sections of an aggregate field of view. In some embodiments, the levels of confidence may be determined based on one or more errors associated with an individual sensor, one or more gross-level degradations, one or more fine-level degradations, or one or more occlusions being present in the sensor data. In some embodiments, the method may additionally include performing one or more operations based on the visibility confidence model or the level of confidence corresponding to individual sub-areas of the aggregate field of view.

Abstract: A method, including loading, to a first address in a memory of a computer, an interposing method having an interposing name that is identical to an original name of an original method in a library of original methods that are to be dynamically loaded by a software application running on the computer, and identifying a second address in the memory of a dynamic linking method associated with the library. A request is then intercepted from the application, the request including a specified name for one of the methods in the library. When the specified name matches the interposing name, the application is directed to the first address so as to run the interposing method. However, when the specified name does not match any interposing name, the application is directed to the second address so as to run the dynamic linking method to run one of the original methods.