Existing work in continual learning (CL) focuses on mitigating catastrophic forgetting, ie,
model performance deterioration on past tasks when learning a new task. However, the …

Recently, a new trend of exploring sparsity for accelerating neural network training has
emerged, embracing the paradigm of training on the edge. This paper proposes a novel …

Although deep learning has made great progress in recent years, the exploding economic
and environmental costs of training neural networks are becoming unsustainable. To …

Unstructured pruning is well suited to reduce the memory footprint of convolutional neural
networks (CNNs), both at training and inference time. CNNs contain parameters arranged in …

Deep neural networks (DNNs) are effective in solving many real-world problems. Larger
DNN models usually exhibit better quality (eg, accuracy) but their excessive computation …

State-of-the-art deep learning models have a parameter count that reaches into the billions.
Training, storing and transferring such models is energy and time consuming, thus costly. A …

Recently, sparse training has emerged as a promising paradigm for efficient deep learning
on edge devices. The current research mainly devotes the efforts to reducing training costs …

State-of-the-art deep neural network (DNN) pruning techniques, applied one-shot before
training starts, evaluate sparse architectures with the help of a single criterion-called pruning …

We present ELSA, a practical solution for creating deep networks that can easily be
deployed at different levels of sparsity. The core idea is to embed one or more sparse …

Mitigating the nonlinear weight update of synaptic devices is one of the main challenges in
designing compute-in-memory (CIM) crossbar arrays for artificial neural networks (ANNs) …