Neural Networks Clustering
Neural Networks and Clustering (Autoencoders) youtube
- #1 : 개요, Autoencoder
- #2: Autoencoder + k-means, "autoencode based data slustering", 2013 song, et al
- #3: Autoencoder + Spectral Clustering, "Learning Deep Representation for graph clustering", 2014, tian et al.
- #4: Autoencoders + Sparese Subspace Clustering, "Deep subspace clustering Net, Ji, et al.
AUtoencoder
- 각 레이어는 representation이다.
- 입력과 출력이 같은 네트워크를 만들면
- 네트워크의 중간을 압축(작게)해도 출력을 생성할수 있는 충분한 정보(represeation)를 가지고 있을 것이다.
[기본적 방법] 1. Run autoencoder 2. Get f(x) 3. Run k-mean of f(x)
Neural Networks and Clustering =
- Neural Networks Clustering
참고 자료
Convolutional Autoencoder: Clustering Images with Neural Networks: blog, 2018.03.23
How to do Unsupervised Clustering with Keras : blog, 2018.06
Clustering with Deep Learning: Taxonomy and New Methods: 2018.01 [깃허브]
DeepCluster: A General Clustering Framework based on Deep Learning: 2017
Essentials of Deep Learning: Introduction to Unsupervised Deep Learning (with Python codes): 2018.05, analyticsvidhya