Abstract: Certain aspects of the present disclosure provide a method, including: processing input data with a feature extraction stage of a machine learning model to generate a feature map; applying an attention map to the feature map to generate an augmented feature map; processing the augmented feature map with a refinement stage of the machine learning model to generate a refined feature map; processing the refined feature map with a first regression stage of the machine learning model to generate multi-dimensional task output data; and processing the refined feature data with an attention stage of the machine learning model to generate an updated attention map.

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: A method for pruning weights of an artificial neural network based on a learned threshold includes determining a pruning threshold for pruning a first set of pre-trained weights of multiple pre-trained weights based on a function of a classification loss and a regularization loss. Weights are pruned from the first set of pre-trained weights when a first value of the weight is less than the pruning threshold. A second set of pre-trained weights of the multiple pre-trained weights is fine-tuned or adjusted in response to a second value of each pre-trained weight in the second set of pre-trained weights being greater than the pruning threshold.

Abstract: Certain aspects of the present disclosure provide techniques for using shaped convolution kernels, comprising: receiving an input data patch, and processing the input data patch with a shaped kernel to generate convolution output.

Abstract: Certain aspects of the present disclosure provide techniques for kernel expansion. An input data tensor is received at a first layer in a neural network, and a first convolution is performed for a first kernel, where the first kernel has a size greater than a preferred size. Performing the first convolution comprises generating a plurality of intermediate tensors by performing a plurality of intermediate convolutions using a plurality of intermediate kernels with a size of the preferred size, and accumulating the plurality of intermediate tensors to generate an output tensor for the first convolution.

Abstract: Certain aspects of the present disclosure provide a method, including: processing input data with a feature extraction stage of a machine learning model to generate a feature map; applying an attention map to the feature map to generate an augmented feature map; processing the augmented feature map with a refinement stage of the machine learning model to generate a refined feature map; processing the refined feature map with a first regression stage of the machine learning model to generate multi-dimensional task output data; and processing the refined feature data with an attention stage of the machine learning model to generate an updated attention map.

Abstract: Certain aspects of the present disclosure provide techniques for provide a method, comprising: receiving input data for a layer of a neural network model; selecting a target code for the input data; and determining weights for the layer based on an autoencoder loss and the target code.

Abstract: Certain aspects of the present disclosure provide techniques for improved machine learning using gradient pruning, comprising computing, using a first batch of training data, a first gradient tensor comprising a gradient for each parameter of a parameter tensor for a machine learning model; identifying a first subset of gradients in the first gradient tensor based on a first gradient criteria; and updating a first subset of parameters in the parameter tensor based on the first subset of gradients in the first gradient tensor.

Abstract: Certain aspects of the present disclosure provide techniques for performing machine learning, including generating a set of basis masks for a convolution layer of a machine learning model, wherein each basis mask comprises a binary mask; determining a set of scaling factors, wherein each scaling factor of the set of scaling factors corresponds to a basis mask in the set of basis masks; generating a composite kernel based on the set of basis masks and the set of scaling factors; and performing a convolution operation based on the composite kernel.

Abstract: A method for pruning weights of an artificial neural network based on a learned threshold includes designating a group of pre-trained weights of an artificial neural network to be evaluated for pruning. The method also includes determining a norm of the group of pre-trained weights, and performing a process based on the norm to determine whether to prune the entire group of pre-trained weights.

Abstract: A method for pruning weights of an artificial neural network based on a learned threshold includes determining a pruning threshold for pruning a first set of pre-trained weights of multiple pre-trained weights based on a function of a classification loss and a regularization loss. The first set of pre-trained weights is pruned in response to a first value of each pretrained weight in the first set of pre-trained weights being greater than the pruning threshold. A second set of pre-trained weights of the multiple pre-trained weights is fine-tuned or adjusted in response to a second value of each pre-trained weight in the second set of pre-trained weights being greater than the pruning threshold.