Self-supervised learning learns representations from the physical nature Self-supervised visual representation learning is a method that learns the physical essence of nature objectives. It is a promising subclass of unsupervised learning. 2022-05-29 Machine learning Deep learning self-supervised nature
Word embedding - Class review We use word embedding to feature representation. The contents of the blog are a note rearranging of course Sequence Model. 2022-05-26 Machine learning Deep learning word embedding representations
Unsupervised learning can learn from features without label Unsupervised learning can learn features by specifically designed loss without any pre-labels. There are two most common seen unsupervised learning methods, which are autoencoder and GANs. In this art 2022-05-18 Machine learning Deep learning unsupervised learning
Weakly supervised learning has various diversities of labels Weak supervision has three types, incomplete supervision, inexact supervision, and inaccurate supervision. Weakly supervised learning is generally defined as a learning framework under inadequate supe 2022-05-17 Machine learning Deep learning weakly-supervised learning
One server can collaborate with global clients using federated learning Live data of humans and other natural mechanisms are constantly generated every day. With the development of human living standards and perceptron, new types of data will be updated. Thus, there is a 2022-05-15 Machine learning Deep learning federated learning federated averaging
Semi-supervised learning and its goal Semi-supervised learning is a learning paradigm concerned with how to learn the presence of both labelled and unlabelled data. 2022-05-14 Machine learning Deep learning semi-supervised
Let's register images to demonstrate their features Image registration can bring two or more images into spatial correspondence. Researchers try their utmost to make or adjust images as as to make them correspond exactly. 2022-05-12 Machine learning Deep learning image registration
Single Shot Detection (SSD) - Lite Version SSDLite is widely used to demonstrate a new backbone's detection capability. It was first introduced in MobileNetV2. 2022-05-12 Machine learning Deep learning detection SSDLite
The evolution of DeepLab semantic segmentation system Deep labelling (DeepLab) is one of the most popular tricks to improve semantic segmentation performance. It may be the most accepted method to improve your proposed method performance. 2022-05-11 Machine learning Deep learning deepLab semantic segmentation
hexo 博客創建和模板 Dave經過一年多博客的撰寫營運,積累了一些寫hexo博客的技巧,並且有自己的一套模板,這裡分享一下。 2022-04-24 HTML blogs generation template