Abstract: A method including processing a stream of data in a sequence of tasks. The processing including receiving a first block of data of the stream of data, determining features associated with the first block of data, selecting, based on the features, one of a first a task to process the first block of data or a second task to process the first block of data and if the second task is selected, shift an output of the second task in time to align the output of the second task with a predicted output of the first task processing a second block of data of the stream of data.

Abstract: Certain aspects of the present disclosure provide techniques for training and using machine learning models to predict locations of stationary and non-stationary objects in a spatial environment. An example method generally includes measuring, by a device, a plurality of signals within a spatial environment. Timing information is extracted from the measured plurality of signals. Based on a machine learning model, the measured plurality of signals within the spatial environment, and the extracted timing information, locations of stationary reflection points and locations of non-stationary reflection points in the spatial environment are predicted. One or more actions are taken by the device based on predicting the locations of stationary reflection points and non-stationary reflection points in the spatial environment.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for training a machine learning model. An example method generally includes partitioning a machine learning model into a plurality of partitions. A request to update a respective partition of the plurality of partitions in the machine learning model is transmitted to each respective participating device of a plurality of participating devices in a federated learning scheme, and the request may specify that the respective partition is to be updated based on unique data at the respective participating device. Updates to one or more partitions in the machine learning model are received from the plurality of participating devices, and the machine learning model is updated based on the received updates.

Abstract: Systems and techniques are described herein for performing optical flow estimation for one or more frames. For example, a process can include determining an optical flow prediction associated with a plurality of frames. The process can include determining a position of at least one feature associated with a first frame and determining, based on the position of the at least one feature in the first frame and the optical flow prediction, a position estimate of a search area for searching for the at least one feature in a second frame. The process can include determining, from within the search area, a position of the at least one feature in the second frame.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for feature masking. A feature tensor is accessed in a neural network, and a feature mask is generated by processing the feature tensor using a masking subnetwork, where the masking subnetwork was trained based at least in part on a polarization constraint and an activation constraint to generate feature masks. A masked feature tensor is generated based on the feature tensor and the feature mask, and an output inference is generated using the neural network based at least in part on the masked feature tensor.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for surrogated federated learning. A set of intermediate activations is received at a trusted server from a node device, where the node device generated the set of intermediate activations using a first set of layers of a neural network. One or more weights associated with a second set of layers of the neural network are refined using the set of intermediate activations, and one or more weight updates corresponding to the refined one or more weights are transmitted to a federated learning system.

Abstract: Certain aspects of the present disclosure provide techniques for desparsified convolution. A weight tensor having unstructured sparsity is accessed, and a densified weight tensor is generated based on the weight tensor by directionally squeezing the weight tensor to remove sparse values, and generating a sparsity map based on the directional squeezing. The densified weight tensor and sparsity map are output for use in a convolutional neural network.

Abstract: Certain aspects of the present disclosure provide techniques for desparsified convolution. An activation tensor is received, and a convolution output is generated for the activation tensor, comprising: selecting a subset of weight elements, corresponding to a set of non-zero elements in the activation tensor, from a weight tensor, and multiplying the set of non-zero elements and the set of weight elements.

Abstract: Methods, systems, and devices for wireless communications are described. A transmitter may transmit control signaling indicating a common resource commonly allocated to a set of receivers, a dedicated resource allocated to a first receiver, and a ramification coding structure that indicates a common code segment and a dedicated code segment of a ramification codeword. The transmitter may determine the ramification coding structure based on common data for multiple receivers, a quantity of hierarchical levels, a machine learning model, or any combination thereof. The transmitter may encode common and dedicated data into at least one ramification codeword in a dynamic signaling format using the ramification coding structure, and transmit the at least one ramification codeword to the first receiver via the common and dedicated resources. Being aware of the ramification coding structure, the first receiver may decode the ramification codeword transmitted by the transmitter in the dynamic signaling format.

Abstract: Certain aspects of the present disclosure provide techniques for machine learning-based deblurring. An input image is received, and a deblurred image is generated based on the input image using a neural network, comprising: generating a feature tensor by processing the input image using a first portion of the neural network, generating a motion mask by processing the feature tensor using a motion portion of the neural network, and generating the deblurred image by processing the feature tensor and the motion mask using a deblur portion of the neural network.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for messaging in a wireless communications system using neural networks. An example method generally includes receiving a first message and a second message, wherein the second message comprises a secret message to be hidden in the first message. The first message and second message are combined into a combined message. An emulation message is generated through an encoder neural network based on the combined message. The emulation message generally comprises a message decodable by a receiving device into the first message. The emulation message is output emulation message for transmission to the receiving device.

Abstract: Disclosed are systems, methods, and non-transitory media for performing radio frequency sensing detection operations. For instance, radio frequency data can be received that is associated with at least one wireless device. The radio frequency data can be based on radio frequency signals reflected from a first object and received by the at least one wireless device. Training label data can also be obtained (e.g., from a labeling device, from the at least one wireless device, etc.). The training label data can be based at least in part on the first object and input data (e.g., received by the labeling device, the at least one wireless device, etc.). A sensing model can be generated based on the radio frequency data and the training label data.

Abstract: An apparatus may utilize an air interface to transmit and/or receive a transmission during a first TTI that includes a second set of data overriding a first set of data scheduled for transmission during the first TTI. The air interface may further be utilized to transmit and/or receive a transmission after a duration of time that is less than a total time duration of the first TTI that includes an override indicator that is at least partially embedded in a data portion of a subframe. The override indicator may be configured to indicate that the first set of data scheduled for transmission during the first TTI is overridden by the second set of data having the higher priority. The override indicator may be transmitted after the second set of data is transmitted. The one or more additional TTIs may be after the first TTI.

Abstract: Certain aspects of the present disclosure provide techniques for efficient bottleneck processing via dimensionality transformation. The techniques include receiving a tensor, and processing the tensor in a bottleneck block in a neural network model, comprising applying a space-to-depth tensor transformation, applying a depthwise convolution, and applying a depth-to-space tensor transformation.

Abstract: Certain aspects of the present disclosure provide techniques for generating fine depth maps for images of a scene based on semantic segmentation and segment-based refinement neural networks. An example method generally includes generating, through a segmentation neural network, a segmentation map based on an image of a scene. The segmentation map generally comprises a map segmenting the scene into a plurality of regions, and each region of the plurality of regions is generally associated with one of a plurality of categories. A first depth map of the scene is generated through a first depth neural network based on a depth measurement of the scene. A second depth map of the scene is generated through a depth refinement neural network based on the segmentation map and the first depth map. One or more actions are taken based on the second depth map of the scene.

Abstract: Systems and techniques are described herein for performing optical flow estimation for one or more frames. For example, a process can include determining an optical flow prediction associated with a plurality of frames. The process can include determining a position of at least one feature associated with a first frame and determining, based on the position of the at least one feature in the first frame and the optical flow prediction, a position estimate of a search area for searching for the at least one feature in a second frame.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for efficient processing of visual content using machine learning models. An example method generally includes generating, from an input, an embedding tensor for the input. The embedding tensor for the input is projected into a reduced-dimensional space projection of the embedding tensor based on a projection matrix. An attention value for the input is derived based on the reduced-dimensional space projection of the embedding tensor and a non-linear attention function. A match, in the reduced-dimensional space, is identified between a portion of the input and a corresponding portion of a target against which the input is evaluated based on the attention value for the input. One or more actions are taken based on identifying the match.

Abstract: A computer-implemented method includes receiving a first input. The first input is interpolated based on a first shift along a first dimension and a second shift along a second dimension. A first output is generated based on the interpolated first input. The first output corresponds to a vectorized bilinear shift of the first input for use in place of grid sampling algorithms.

Abstract: Embodiments described herein relate to a method, comprising: receiving input data at a convolutional neural network (CNN) model; generating a factorized computation network comprising a plurality of connections between a first layer of the CNN model and a second layer of the CNN model, wherein: the factorized computation network comprises N inputs, the factorized computation network comprises M outputs, and the factorized computation network comprises at least one path from every input of the N inputs to every output of the M outputs; setting a connection weight for a plurality of connections in the factorized computation network to 1 so that a weight density for the factorized computation network is <100%; performing fast pointwise convolution using the factorized computation network to generate fast pointwise convolution output; and providing the fast pointwise convolution output to the second layer of the CNN model.

Abstract: Systems and techniques are described herein for performing optical flow estimation for one or more frames. For example, a process can include determining an optical flow prediction associated with a plurality of frames. The process can include determining a position of at least one feature associated with a first frame and determining, based on the position of the at least one feature in the first frame and the optical flow prediction, a position estimate of a search area for searching for the at least one feature in a second frame. The process can include determining, from within the search area, a position of the at least one feature in the second frame.