Wrong. Neural networks are affinity invariant. You can rotate, skew, translate t... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

wookietrader on April 19, 2013 | parent | context | favorite | on: “‘Principal Component Analysis’ is a dimensionally...

Wrong.

Neural networks are affinity invariant. You can rotate, skew, translate the input data however you want, the optimum stays the same. Same for SVMs.

tgflynn on April 19, 2013 [–]

But the weights will be different, which it seems to me is comparable to saying that the principal components change.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact