I’ve been working with SVDs a lot in my research recently, and have gotten interested in manifold learning as a result.
What are your data sets like and how well has your manifold learning algorithms worked on them? I get the feeling that, for instance, Local Linear Embedding wouldn’t do so well factoring out rotation in computer vision.
Awesome, thanks for posting this.
I’ve been working with SVDs a lot in my research recently, and have gotten interested in manifold learning as a result.
What are your data sets like and how well has your manifold learning algorithms worked on them? I get the feeling that, for instance, Local Linear Embedding wouldn’t do so well factoring out rotation in computer vision.