Exploring the Potential of Data: A Journey of Representation Learning with Pre-trained Models and Masked Autoencoder
Original link: https://controlnet.space/2023/06/08/reading/representation-learning/ Representation Learning (Representation Learning) is a concept in deep learning. By pre-training a feature extractor, the original data is converted into meaningful low-dimensional features, and downstream tasks are trained based on these features, thereby reducing the need for data and computing. capacity needs. This article will introduce the basic concepts of representation […]