class07 (1)

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

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

Stability of Tikhonov Regularization 9.520 Class 07, March 2003 Alex Rakhlin

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

View Full Document
Plan Review of Stability Bounds Stability of Tikhonov Regularization Algorithms
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 implies good generalization bounds. This class : Tikhonov Regularization has uniform stability of β = O 1 . 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

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

View Full Document
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 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 ) .
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 differently, 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 ( · ) |

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

View Full Document
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 difference 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 -insensitive loss are both L -Lipschitz with L = 1. The square loss function is L Lipschitz if we can bound the y values and the f ( x ) values generated. The 0 - 1 loss function is not L
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern