Fenchel-Moreau and Primal/Dual Optimization Problems: Difference between revisions

From Optimal Transport Wiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 9: Line 9:
Let''X'' be a normed vector space, and let ''X*'' denote its topological dual. Given an extended real-valued function <math> f: X \to \mathbb{R} \cup \{+\infty\} </math>, its ''convex conjugate'' <math>f^*:X^* \to \mathbb{R} \cup \{+\infty\}</math> is defined by  
Let''X'' be a normed vector space, and let ''X*'' denote its topological dual. Given an extended real-valued function <math> f: X \to \mathbb{R} \cup \{+\infty\} </math>, its ''convex conjugate'' <math>f^*:X^* \to \mathbb{R} \cup \{+\infty\}</math> is defined by  


:<math> f^*(u):=\sup_{x \in X} \{ \langle u,x \rangle - f(x) \} \quad \forall u \in X^*. </math>
:<math> f^*(y):=\sup_{x \in X} \{ \langle y,x \rangle - f(x) \} \quad \forall y \in X^*. </math>


An immediate consequence of this definition is ''Young's Inequality'',
An immediate consequence of this definition is ''Young's Inequality'',


:<math> f^*(u) +f(x) \geq \langle u,x \rangle \quad \forall x \in X, u \in X^* . </math>
:<math> f^*(y) +f(x) \geq \langle y,x \rangle \quad \forall x \in X, y \in X^* . </math>


Furthermore, it follows directly from the definition that, for ''any'' function ''f'', its conjugate function ''f*'' is convex and lower semicontinuous.
Furthermore, it follows directly from the definition that, for ''any'' function ''f'', its conjugate function ''f*'' is convex and lower semicontinuous.
Line 19: Line 19:
In a similar way, for any function ''f'', its the ''biconjugate'' function <math>f^{**}:X \to \mathbb{R} \cup \{+\infty\} </math> is defined by  
In a similar way, for any function ''f'', its the ''biconjugate'' function <math>f^{**}:X \to \mathbb{R} \cup \{+\infty\} </math> is defined by  


:<math> f^{**}(x):=\sup_{u \in X^*} \{ \langle u,x \rangle - f^*(u) \} \quad \forall u \in X^*. </math>
:<math> f^{**}(x):=\sup_{y \in X^*} \{ \langle y,x \rangle - f^*(y) \} \quad \forall y \in X^*. </math>


As above, for any function ''f'', its biconjugate function ''f**'' is convex and lower semicontinuous. Furthermore, by Young's inequality, we always have
As above, for any function ''f'', its biconjugate function ''f**'' is convex and lower semicontinuous. Furthermore, by Young's inequality, we always have
Line 28: Line 28:


   
   
==Application to Primal/Dual Optimization Problems==
An important consequence of the Fenchel-Moreau Theorem is characterizing the equivalence of the primal and dual optimization problems.<ref name="Rockafellar" /> Given a normed vector space ''X'' and a proper, lower semicontinuous, convex function <math>f: X \to \mathbb{R}\cup \{+\infty\}</math>, the ''primal'' optimization problem is given by
 
:<math> \inf_{x \in X} f(x). </math>
 
The corresponding dual problem arises from a suitable ``perturbation'' of the primal problem, subject to a parameter ''u'' ∈ ''U'', where ''U'' is also a normed vector space. In particular, let <math> F:X \times U \to \mathbb{R} \cup \{+\infty\} </math> be a proper convex function so that <math> f(x) = F(x,0) </math>. Then the ''primal'' and ''dual'' problems may be written as
 
:<math> P_0 := \inf_{x \in X} F(x,0) </math>
:<math> D_0 := \sup_{v \in U^*} -F^*(0,v) </math>
 
 


==References==
==References==
<references>
<references>
<ref name="Brezis">H. Brezis, ''Functional Analysis'', Chapter 1.</ref>
<ref name="Brezis">H. Brezis, ''Functional Analysis'', Chapter 1.</ref>
<ref name="Brezis">Rockafellar, ''Convex Analysis''.</ref>
</references>
</references>

Revision as of 22:54, 7 April 2020

The Fenchel-Moreau Theorem[1] is a fundamental result in convex analysis, characterizing the class of functions for which a function equals its biconjugate. A key consequence of this theorem is the equivalence of primal and dual optimization problems.

Fenchel-Moreau Theorem

Given a normed vector space X and , then

Background on Convex Conjugate Functions

LetX be a normed vector space, and let X* denote its topological dual. Given an extended real-valued function , its convex conjugate is defined by

An immediate consequence of this definition is Young's Inequality,

Furthermore, it follows directly from the definition that, for any function f, its conjugate function f* is convex and lower semicontinuous.

In a similar way, for any function f, its the biconjugate function is defined by

As above, for any function f, its biconjugate function f** is convex and lower semicontinuous. Furthermore, by Young's inequality, we always have

Since f** is always convex and lower semicontinuous, in order for equality to hold, it is necessary for f to be convex and lower semicontinuous. The heart of Fenchel-Moreau Theorem is that this condition is also sufficient.


Application to Primal/Dual Optimization Problems

An important consequence of the Fenchel-Moreau Theorem is characterizing the equivalence of the primal and dual optimization problems.[2] Given a normed vector space X and a proper, lower semicontinuous, convex function , the primal optimization problem is given by

The corresponding dual problem arises from a suitable ``perturbation of the primal problem, subject to a parameter uU, where U is also a normed vector space. In particular, let be a proper convex function so that . Then the primal and dual problems may be written as


References

  1. H. Brezis, Functional Analysis, Chapter 1. Cite error: Invalid <ref> tag; name "Brezis" defined multiple times with different content
  2. Cite error: Invalid <ref> tag; no text was provided for refs named Rockafellar