Deep neural networks (DNNs) have shown great success in completing complex tasks.
However, DNNs inevitably bring high computational cost and storage consumption due to …

Low-precision deep neural network (DNN) training has gained tremendous attention as
reducing precision is one of the most effective knobs for boosting DNNs' training time/energy …

State-of-the-art visual place recognition performance is currently being achieved utilizing
deep learning based approaches. Despite the recent efforts in designing lightweight …

Low precision training can significantly reduce the computational overhead of training deep
neural networks (DNNs). Though many such techniques exist, cyclic precision training …

Contrastive learning learns visual representations by enforcing feature consistency under
different augmented views. In this work, we explore contrastive learning from a new …

Recently, significant improvements have been achieved for object detection algorithm by
increasing the size of convolutional neural network (CNN) models, but the resulting increase …

Digital life, a form of life generated by computer programs or artificial intelligence systems, it
possesses self-awareness, thinking abilities, emotions, and subjective consciousness …

Since not all data can be represented in Euclidean space [26], many applications rely on
graph-structured data. For example, social networks can be modeled as graphs by …

The use of mixed precision values when training an artificial neural network (ANN) can
increase performance while reducing cost. Certain portions and/or steps of an ANN may be …

Quantization is known as an effective compression method for deploying neural networks on
mobile devices. However, most existing works train from scratch a quantized network with …