I'm familiar with Auto-Encoders and I'm about to dive into CNNs. By having a look at the most important component of a CNN, the filter:
I wonder how it is different from Auto-Encoders:
For me, it looks conceptually the same. I even have to admit it is quite of another higher concept: Dimensionality reduction. In e.g. PCA, as well as in AE and in CNN(?) you transform higher data / higher dimensions onto lower / compressed data. I can see that the methods are somehow different but yet, can't really explain in which manner, finally.