Comparison of different autoencoder variations vs state-of-the-art methods on Sim1 (17 clusters), the optimal value of neighbours for Isomap was found as 110 through grid search.
收藏数据链接：
官方服务：
资源简介：
The metrics indicate that several variations of autoencoders and the Isomap algorithm are able to outperform conventional algorithms such as PCA or ICA for the Sim1 dataset that contains a high number of clusters. Moreover, two autoencoder variants outperform Isomap on 5 out of the 6 metrics, the values indicating the highest performance for each metric have been bolded.
创建时间：
2023-03-09
