这里收集了一些在阅读 paper 时遇到的好句子。

Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection

---

In this paper, we propose Deep Autoencoding Gaussian Mixture Model (DAGMM), a deep learning framework that addresses the aforementioned challenges in unsupervised anomaly detection from several aspects.

• aforementioned adj.上述的；前面提及的

---

Usually, it is hard to learn deep autoencoders by end-to-end training, as they can be easily stuck in less attractive local optima, so pre-training is widely adopted.

• be easily stuck in less attractive local optima：陷入不那么吸引人的局部最优状态

---

It is hard to make any significant change to a well-trained autoencoder via fine-tuning.

---

Given a dataset of N samples, the objective function that guides DAGMM training is constructed as follows.

• 用于引出这个 model 的 objective function

---