Abstract: Systems and techniques are described for adaptive mixed-resolution processing. According to some aspects, a device can divide an input image into first tokens having a first resolution and second tokens having a second resolution. The device can generate first token representations for token(s) from the first tokens corresponding to a first region of the input image and generate second token representations for token(s) from the second tokens corresponding to the first region of the input image. The device can process, using a neural network model, the first token representations and the second token representations to determine the first resolution or the second resolution as a scale for the first region of the input image. The device can process, using a transformer neural network model, the first region of the input image according to the scale for the first region.

Abstract: A computing device may be configured to intelligently activate gating within a current layer of a neural network that includes two or more filters. The computing device may receive a layer-specific input data that is specific to the current layer of the neural network, generate statistics based on the received layer-specific input data; and use the generated statistics to assign a relevance score to each of the two or more filters. Each assigned relevance score may indicate the relevance of the corresponding filter to the received layer-specific input data. The computing device may determine an activation status of each of the two or more filters in the current layer based on the identified relevance and apply the received layer-specific input data to the activated filters in the two or more filters to generate an output activation for the current layer of the neural network.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for machine learning. In one aspect, base model output data is generated, the generating including processing input data with at least a portion of a base model of a machine learning model architecture, and the base model output data is processed with a routing model of the machine learning model architecture in order to determine a selected expert model, of a plurality of expert models, with which to process the base model output data. Expert model output data is generated, where generating the expert model output data includes processing the base model output data with the selected expert model, and final output data from the machine learning model architecture is generated, where generating the final output data includes processing the base model output data and the expert model output data with an ensemble model of the machine learning model architecture.

Abstract: A processor-implemented method of video processing using includes receiving, via an artificial neural network (ANN), a video including a first frame and a second frame. A saliency map is generated based on the first frame of the video. The second frame of the video is sampled based on the saliency map. A first portion of the second frame is sampled at a first resolution and a second portion of the second frame is sampled at a second resolution. The first resolution is different than the second resolution. A resampled second frame is generated based on the sampling of the second frame. The resampled second frame is processed to determine an inference associated with the video.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for processing a video stream using a machine learning model. An example method generally includes generating a first group of tokens from a first frame of the video stream and a second group of tokens from a second frame of the video stream. A first set of tokens associated with features to be reused from the first frame and a second set of tokens associated with features to be computed from the second frame are identified based on a comparison of tokens from the first group of tokens to corresponding tokens in the second group of tokens. A feature output is generated for portions of the second frame corresponding to the second set of tokens. Features associated with the first set of tokens are combined with the generated feature output into a representation of the second frame.

Abstract: Certain aspects of the present disclosure provide techniques for processing with an auto exiting machine learning model architecture, including processing input data in a first portion of a classification model to generate first intermediate activation data; providing the first intermediate activation data to a first gate; making a determination by the first gate whether or not to exit processing by the classification model; and generating a classification result from one of a plurality of classifiers of the classification model.

Abstract: A method for video processing via an artificial neural network includes receiving a video stream as an input at the artificial neural network. A residual is computed based on a difference between a first feature of a current frame of the video stream and a second feature of a previous frame of the video stream. One or more portions of the current frame of the video stream are processed based on the residual. Additionally, processing is skipped for one or more portions of the current frame of the video based on the residual.

Abstract: Various aspects provide methods for learning, such as continual learning, that support task-incremental learning using a multi-head classification architecture. Various aspects may enable conditional computing to support multi-head classification. Various aspects provide methods for learning, such as continual learning, that support class-incremental learning using a single-head classification architecture. Various aspects may enable conditional computing to support single-head classification by predicting the task associated with a given test input and selecting an associated classification head based at least in part on the task prediction.

Abstract: A computing device may be equipped with a generalized framework for accomplishing conditional computation or gating in a neural network. The computing device may receive input in a neural network layer that includes two or more filters. The computing device may intelligently determine whether the two or more filters are relevant to the received input. The computing device may deactivate filters that are determined not to be relevant to the received input (or activate filters that are determined to be relevant to the received input), and apply the received input to active filters in the layer to generate an activation.

Abstract: An embodiment of a method to characterize a die is disclosed. The embodiment of the method includes measuring a quality metric of the die, and determining, prior to a final test stage, whether the quality metric of the die satisfies a first constraint, where the first constraint is more stringent than a second constraint at the final test stage for the quality metric of the die.

Abstract: Methods of using one of a plurality of configuration bitstreams in an integrated circuit are disclosed. An exemplary method comprises analyzing the plurality of implementations of a design to determine initial variations in timing among the implementations; modifying the implementations to reduce the variations in timing among the implementations; and outputting a plurality of configuration bitstreams for the implementations having variations in timing that are reduced relative to the initial variations in timing. Another method comprises generating a plurality of implementations for the design; generating a cost function for the design based upon costs (e.g.