class07 (1)

class07 (1) - Stability of Tikhonov Regularization 9.520...

Info iconThis preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
Stability of Tikhonov Regularization 9.520 Class 07, March 2003 Alex Rakhlin
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Plan Review of Stability Bounds Stability of Tikhonov Regularization Algorithms
Background image of page 2
Uniform Stability Review notation : S = { z 1 , ..., z } ; S i,z = { z 1 , ..., z i - 1 , z, z i +1 , ..., z } c ( f, z ) = V ( f ( x ) , y ), where z = ( x , y ). An algorithm A has uniform stability β if ( S, z ) ∈ Z +1 , i, sup u ∈Z | c ( f S , u ) - c ( f S i,z , u ) | ≤ β. Last class : Uniform stability of β = O ± 1 ² implies good generalization bounds. This class : Tikhonov Regularization has uniform stability of β = O ± 1 ² . Reminder : The Tikhonov Regularization algorithm: f S = arg min f ∈H 1 X i =1 V ( f ( x i ) , y i ) + λ k f k 2 K
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Generalization Bounds Via Uniform Stability If β = k for some k , we have the following bounds from the last lecture: P ± | I [ f S ] - I S [ f S ] | ≥ k + ± ² 2 exp - ‘± 2 2( k + M ) 2 ! . Equivalently, with probability 1 - δ , I [ f S ] I S [ f S ] + k + (2 k + M ) s 2 ln(2 ) .
Background image of page 4
Lipschitz Loss Functions, I We say that a loss function (over a possibly bounded do- main X ) is Lipschitz with Lipschitz constant L if y 1 , y 2 , y 0 ∈ Y , | V ( y 1 , y 0 ) - V ( y 2 , y 0 ) | ≤ L | y 1 - y 2 | . Put diFerently, if we have two functions f 1 and f 2 , under an L -Lipschitz loss function, sup ( x ,y ) | V ( f 1 ( x ) , y ) - V ( f 2 ( x ) , y ) | ≤ L | f 1 - f 2 | . Yet another way to write it: | c ( f 1 , · ) - c ( f 2 , · ) | L | f 1 ( · ) - f 2 ( · ) |
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Lipschitz Loss Functions, II If a loss function is L -Lipschitz, then closeness of two func- tions (in L norm) implies that they are close in loss. The converse is false — it is possible for the diFerence in loss of two functions to be small, yet the functions to be far apart (in L ). Example: constant loss. The hinge loss and the
Background image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 24

class07 (1) - Stability of Tikhonov Regularization 9.520...

This preview shows document pages 1 - 7. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online