Open dkirkby opened 6 years ago
In the first instance, a PR with that comment, and perhaps an illustration of brittleness to parameters, would be welcome
I can submit a PR to update the existing example with document of its inherent sensitivity to small changes in the data and hyperparams. A more robust example would be even better, though, if anyone has ideas.
Quickly playing with the example and tweaking the parameters, it looks like a smaller factor (0.25 for examle) makes the example more robust.
The only text accompanying this kernel PCA example is:
However, this example is very fine tuned and probably gives most readers a misleading impression. To demonstrate this, the following function reproduces the example's projection to a 2D latent space (lower left plot):
With the defaults args, the example is reproduced exactly and the separation is very clear:
However, small changes to any of these args reveal that this nice result is not at all typical, and you are more likely to find a latent space with a similar nonlinearity to the original data:
To improve the pedagogical value of this example, I suggest either finding a more robust demonstration of kernel PCA or else commenting that some fine tuning is generally required to achieve linear separation.