TL;DR提出了一种无监督生成聚类方法Variational Deep Embedding (VaDE),使用高斯混合模型和神经网络来建模数据生成过程,并在VaDE中使用变分推断实现更好的聚类效果,并可生成高度逼真的样本,更广泛的混合模型也可以轻松集成。
Abstract
clustering is among the most fundamental tasks in computer vision and machine learning. In this paper, we propose Variational Deep Embedding (VaDE), a novel unsupervised generative clustering approach within the