search

chengsoonong / mclass-sky

Multiclass methods for astronomical data
BSD 3-Clause "New" or "Revised" License
9 stars 4 forks source link

Fix writeup on kernelised linear regression #179

Closed nbgl closed 7 years ago

nbgl commented 7 years ago
nbgl commented 7 years ago

Completed in fdb1e83.

chengsoonong commented 7 years ago

Found a bug in your derivation.

f = phi(x)^\top w and in your least squares error you say f = Kw

The second is not correct. To go from the first to the second, you need to know that K = \phi^\top \phi The kernel trick: k(x_i, x_j) = < \phi(x_i) , \phi(x_j) >

nbgl commented 7 years ago

Finished in 032cca0 (#181)