We contend that all key features and structures of modern deep (convolution) neural networks can be naturally derived from optimizing a principled objective, namely the rate reduction recently proposed by Yu et al. (2020)
Yu et al. (2020)
Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate Reduction
主要目标:reformulating the objective towards learning explicitly meaningful representations for the data x。把深度学习的目标改为去找一个可以准确解释数据的描述数据的特征向量。