{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# Module2_2 - Module 2 Lecture 2 Fundamental Concepts Entropy...

This preview shows pages 1–8. Sign up to view the full content.

Module 2, Lecture 2 Fundamental Concepts: Entropy, Relative Entropy, Mutual Information G.L. Heileman Module 2, Lecture 2

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

View Full Document
Entropy Definition (Entropy) The entropy of a RV X p ( x ) is: H ( X ) = x ∈X p ( x ) log 1 p ( x ) = E log 1 p ( X ) Properties of H ( X ): 1 H ( X ) 0 with equality when p ( x ) = 1 for some one x ∈ X . 2 H ( X ) log( |X| ) with equality when p ( x ) = 1 |X| x ∈ X . 3 H b ( X ) = (log b 2) H ( X ). Proof: 1 0 p ( x ) 1 log 1 p ( x ) 0. 2 we’ll prove this momentarily. 3 log b c = (log b a )(log a c ). G.L. Heileman Module 2, Lecture 2
Binary Entropy Function p 0 1/2 1 0 1 H(p,1-p) G.L. Heileman Module 2, Lecture 2

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

View Full Document
Entropy Theorem H ( X ) log( |X| ) with equality when p ( x ) = 1 |X| x ∈ X . Proof: Consider H ( 1 n , . . . , 1 n ) - H ( p 1 , . . . , p n ) where the p i ’s are arbitrary probabilities satisfying the probability axioms. We will show that the above quantity is 0 for any valid choices of the p i ’s. To do this, rewrite the previous equation as: log b n + n i =1 p i log b p i = log b e · log e n + n i =1 p i (log e p i · log b e ) = log b e · (ln n + n i =1 p i ln p i ) G.L. Heileman Module 2, Lecture 2
Entropy Proof (con’t): = log b e · (ln n · n i =1 p i + n i =1 p i ln p i ) = log b e · n i =1 p i ln( np i ) . Now, using the fact that ln 1 x 1 - x x > 0, we can write H 1 n , . . . , 1 n - H ( p 1 , . . . , p n ) log b e · n i =1 p i 1 - 1 np i = log b e · n i =1 p i - n i =1 1 n = 0 . Thus, equiprobable events produce maximum entropy. G.L. Heileman Module 2, Lecture 2

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

View Full Document
Joint Entropy Definition (Joint Entropy) The joint entropy of a pair of RVs ( X , Y ) p ( x , y ) is: H ( X , Y ) = x ∈X y ∈Y p ( x , y ) log 1 p ( x , y ) = E log 1 p ( X , Y ) When X and Y are independent, p ( x , y ) = p ( x ) p ( y ), and H ( X , Y ) = x , y p ( x ) p ( y ) log 1 p ( x ) + log 1 p ( y ) = y p ( y ) x p ( x ) log 1 p ( x ) + x p ( x ) y p ( y ) log 1 p ( y ) = H ( X ) + H ( Y ) G.L. Heileman Module 2, Lecture 2
Conditional Entropy The conditional entropy of X given that Y = y is observed can be written as: H ( X | Y = y ) = x p ( x | Y = y ) log 1 p ( x | Y = y ) I.e., this is the entropy of the probability distribution p ( x | Y = y ).

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

View Full Document
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