Unlike ReLU, newer activation functions (like Swish, H-swish, Mish) that are frequently
employed in popular efficient architectures can also result in negative activation values, with …

When training neural networks with simulated quantization, we observe that quantized
weights can, rather unexpectedly, oscillate between two grid-points. The importance of this …

Current model quantization methods have shown their promising capability in reducing
storage space and computation complexity. However, due to the diversity of quantization …

Quantizing deep neural networks is an effective method for reducing memory consumption
and improving inference speed, and is thus useful for implementation in resource …

Deep neural networks with adaptive configurations have gained increasing attention due to
the instant and flexible deployment of these models on platforms with different resource …

Benefiting from tens of millions of hierarchically stacked learnable parameters, Deep Neural
Networks (DNNs) have demonstrated overwhelming accuracy on a variety of artificial …

Abstract Model size and inference speed/power have become a major challenge in the
deployment of neural networks for many applications. A promising approach to address …

Quantization plays an important role in the energy-efficient deployment of deep neural
networks on resource-limited devices. Post-training quantization is highly desirable since it …

We introduce a data-free quantization method for deep neural networks that does not
require fine-tuning or hyperparameter selection. It achieves near-original model …

Quantization is a popular way of increasing the speed and lowering the memory usage of
Convolution Neural Networks (CNNs). When labelled training data is available, network …