Background: Distributed training is essential for large scale training o...
In deep learning, fine-grained N:M sparsity reduces the data footprint a...
Power consumption is a major obstacle in the deployment of deep neural
n...
Despite their growing popularity, graph neural networks (GNNs) still hav...
Quantization of the weights and activations is one of the main methods t...
Recently, researchers proposed pruning deep neural network weights (DNNs...
Recent research has shown remarkable success in revealing "steering"
dir...
Neural gradient compression remains a main bottleneck in improving train...
Neural network quantization methods often involve simulating the quantiz...
Neural network quantization enables the deployment of large models on
re...
Convolutional neural networks (CNNs) have become the dominant neural net...
Convolutional neural networks (CNNs) introduce state-of-the-art results ...
Convolutional neural networks (CNNs) achieve state-of-the-art accuracy i...
Unlike traditional approaches that focus on the quantization at the netw...
Quantized Neural Networks (QNNs) are often used to improve network effic...
Over the past few years batch-normalization has been commonly used in de...