Abstract: A processor-implemented method includes sampling, according to a priority sampling policy, a set of node priorities from a computation graph. Each node priority of the set of node priorities may be associated with a respective node on the computation graph. Additionally, each node may represent an operation of a task performed by an artificial neural network. The method also includes converting, via a list scheduling function, the node priorities to a schedule that associates each node of the computation graph with a processor of a group of processors of a device associated with the artificial neural network, the schedule associated with a makespan. The method further includes performing the task in accordance with the schedule.

Abstract: A processor-implemented method includes generating, by a scheduling model, a group of schedules from a computation graph associated with a task, each node on the computation graph being associated with an operation of an artificial neural network, each schedule of the group of schedules associating each node of the computation graph with a processor of a group of processors of a hardware device. The processor-implemented method also includes testing one or more schedules of the group of schedules on the hardware device or a model of the hardware device. The processor-implemented method further includes selecting a schedule of the one or more schedules based on testing the one or more schedules, the selected schedule satisfying a selection condition.

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.

Abstract: A processor-implemented method for generating a schedule for executing operations of a compute graph includes receiving a graph including multiples nodes connected by edges. Each of the multiple nodes represents an operation to be executed. A set of sequences for executing the nodes is determined based on one or more precedence constraints. One or more sequences are selected from the set of sequences based on a memory constraint associated with a device for executing the nodes. A schedule for executing the nodes on the device is generated based on the selected one or more sequences.

Abstract: A processor-implemented method for generating a topological order using an artificial neural network (ANN) includes receiving a set of tasks to be performed. The tasks are represented in a graph including multiple nodes connected by edges. Each node corresponds to a task in the set of tasks. A scheduling priority is assigned to each node in the graph. A next node of potential next nodes is selected according to a probability of each of the potential next nodes based on the assigned scheduling priorities and a topology of the graph. A topological order of the tasks is generated by repeating the selection of the next node.

Abstract: Certain aspects of the present disclosure provide techniques and apparatus for performing machine learning. In one example, an input data sequence is accessed, and the input data sequence is sliced based on a slice length hyperparameter to generate a stacked slice input data representation. The stacked slice input data representation is processed with a slice attention layer to generate a stacked slice output data representation. The stacked slice output data representation is de-sliced to generate an output data sequence.

Abstract: A method of exploiting activation sparsity in deep neural networks is described. The method includes retrieving an activation tensor and a weight tensor where the activation tensor is a sparse activation tensor. The method also includes generating a compressed activation tensor comprising non-zero activations of the activation tensor, where the compressed activation tensor has fewer columns than the activation tensor. The method further includes processing the compressed activation tensor and the weight tensor to generate an output tensor.

Abstract: A method of exploiting activation sparsity in deep neural networks is described. The method includes retrieving an activation tensor and a weight tensor where the activation tensor is a sparse activation tensor. The method also includes generating a compressed activation tensor comprising non-zero activations of the activation tensor, where the compressed activation tensor has fewer columns than the activation tensor. The method further includes processing the compressed activation tensor and the weight tensor to generate an output tensor.

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.

Abstract: Various aspects provide methods for a computing device selecting a neural network for a hardware configuration including using an accuracy predictor to select from a search space a neural network including a first plurality of the blockwise knowledge distillation trained search blocks, in which the accuracy predictor is built using search space trained blockwise knowledge distillation search blocks. Aspects may include selecting a second plurality of the blockwise knowledge distillation trained search blocks based on criteria of predicted accuracy using the accuracy predictor for the second plurality of the blockwise knowledge distillation trained search blocks.

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.

Abstract: A method performed by a computing device includes determining a partition for depth-first processing by a multi-layer artificial neural network (ANN) of the computing device. The computing device comprising a processor, on-chip memory, and off-chip memory. The first partition determined based on an amount of on-chip memory used by the first partition, an available amount of on-chip memory, and a size of a write back to the off-chip memory. The method also includes processing, at the device via the multi-layer ANN, an input, using the depth-first processing in accordance with the partition.

Abstract: Various embodiments include methods and devices for implementing automatic grammar augmentation for improving voice command recognition accuracy in systems with a small footprint acoustic model. Alternative expressions that may capture acoustic model decoding variations may be added to a grammar set. An acoustic model-specific statistical pronunciation dictionary may be derived by running the acoustic model through a large general speech dataset and constructing a command-specific candidate set containing potential grammar expressions. Greedy based and cross-entropy-method (CEM) based algorithms may be utilized to search the candidate set for augmentations with improved recognition accuracy.

Abstract: A method for defining a sensor model includes determining a probability of obtaining a measurement from multiple potential causes in a field of view of a sensor modeled based on a stochastic map. The stochastic map includes a mean occupancy level for each voxel in the stochastic map and a variance of the mean occupancy level for each pixel. The method also includes determining a probability of obtaining an image based on the determined probability of obtaining the measurement. The method further includes planning an action for a robot, comprising the sensor, based on the probability of obtaining the image.

Abstract: A method of exploiting activation sparsity in deep neural networks is described. The method includes retrieving an activation tensor and a weight tensor where the activation tensor is a sparse activation tensor. The method also includes generating a compressed activation tensor comprising non-zero activations of the activation tensor, where the compressed activation tensor has fewer columns than the activation tensor. The method further includes processing the compressed activation tensor and the weight tensor to generate an output tensor.

Abstract: A method substantially simultaneously plans a path and maps an environment by a robot. The method determines a mean of an occupancy level for a location in a map. The method also includes determining a probability distribution function (PDF) of the occupancy level. The method further includes calculating a cost function based on the PDF. Finally, the method includes simultaneously planning the path and mapping the environment based on the cost function.

Abstract: A diet aid comprises a removable sleeve, having a timing circuit with user-adjustable delay, visual cues, and a slot into which an ordinary eating utensil can be inserted, the insertion of the utensil energizing the circuit, with the delay cycle triggered automatically as the user places a bite of food to their mouth.

Abstract: A method for generating a map includes determining an occupancy level of each of multiple voxels. The method also includes determining a probability distribution function (PDF) of the occupancy level of each voxel. The method further includes performing an incremental Bayesian update on the PDF to generate the map based on a measurement performed after determining the PDF.

Abstract: A method for defining a sensor model includes determining a probability of obtaining a measurement from multiple potential causes in a field of view of a sensor modeled based on a stochastic map. The stochastic map includes a mean occupancy level for each voxel in the stochastic map and a variance of the mean occupancy level for each pixel. The method also includes determining a probability of obtaining an image based on the determined probability of obtaining the measurement. The method further includes planning an action for a robot, comprising the sensor, based on the probability of obtaining the image.

Abstract: A method of calculating a most likely map based on batch data includes gathering a corpus of sensor measurements indexed by a location of a sensor throughout an environment to be mapped. The method also includes determining, after gathering the corpus of sensor measurements, a most likely occupancy level of each voxel of multiple voxels of the environment in accordance with the corpus of sensor measurements and a stochastic sensor model. The method further includes calculating the most likely map based on the determined most likely occupancy level.