Namespaces
Variants
Actions

Difference between revisions of "Gâteaux gradient"

From Encyclopedia of Mathematics
Jump to: navigation, search
m (moved Gateaux gradient to Gâteaux gradient over redirect: accented title)
m (tex encoded by computer)
 
Line 1: Line 1:
''of a functional <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433801.png" /> at a point <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433802.png" /> of a Hilbert space <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433803.png" />''
+
<!--
 +
g0433801.png
 +
$#A+1 = 18 n = 0
 +
$#C+1 = 18 : ~/encyclopedia/old_files/data/G043/G.0403380 G\Awateaux gradient
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
  
The vector in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433804.png" /> equal to the [[Gâteaux derivative|Gâteaux derivative]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433805.png" /> of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433806.png" /> at <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433807.png" />. In other words, the Gâteaux gradient is defined by the formula
+
{{TEX|auto}}
 +
{{TEX|done}}
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433808.png" /></td> </tr></table>
+
''of a functional  $  f $
 +
at a point  $  x _ {0} $
 +
of a Hilbert space  $  H $''
  
where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g0433809.png" /> as <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338010.png" />. In an <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338011.png" />-dimensional Euclidean space the Gâteaux gradient <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338012.png" /> is the vector with coordinates
+
The vector in  $  H $
 +
equal to the [[Gâteaux derivative|Gâteaux derivative]]  $  f _ {G} ^ { \prime } ( x _ {0} ) $
 +
of  $  f $
 +
at  $  x _ {0} $.  
 +
In other words, the Gâteaux gradient is defined by the formula
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338013.png" /></td> </tr></table>
+
$$
 +
f ( x _ {0} + h)  = \
 +
f ( x _ {0} ) +
 +
( f _ {G} ^ { \prime } ( x _ {0} ), h) + \epsilon ( h),
 +
$$
  
and is simply known as the [[Gradient|gradient]]. The concept of the Gâteaux gradient may be extended to the case when <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338014.png" /> is a Riemannian manifold (finite-dimensional) or an infinite-dimensional Hilbert manifold and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338015.png" /> is a smooth real function on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338016.png" />. The growth of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338017.png" /> in the direction of its Gâteaux gradient is larger than in any other direction passing through the point <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/g/g043/g043380/g04338018.png" />.
+
where  $  \epsilon ( th)/t \rightarrow 0 $
 +
as  $  t \rightarrow 0 $.
 +
In an  $  n $-
 +
dimensional Euclidean space the Gâteaux gradient  $  f _ {G} ^ { \prime } ( x _ {0} ) $
 +
is the vector with coordinates
 +
 
 +
$$
 +
\left (
 +
 
 +
\frac{\partial  f ( x _ {0} ) }{\partial  x _ {1} }
 +
\dots
 +
 
 +
\frac{\partial  f ( x _ {0} ) }{\partial  x _ {n} }
 +
\right ) ,
 +
$$
 +
 
 +
and is simply known as the [[Gradient|gradient]]. The concept of the Gâteaux gradient may be extended to the case when $  X $
 +
is a Riemannian manifold (finite-dimensional) or an infinite-dimensional Hilbert manifold and $  f $
 +
is a smooth real function on $  X $.  
 +
The growth of $  f $
 +
in the direction of its Gâteaux gradient is larger than in any other direction passing through the point $  x _ {0} $.

Latest revision as of 19:42, 5 June 2020


of a functional $ f $ at a point $ x _ {0} $ of a Hilbert space $ H $

The vector in $ H $ equal to the Gâteaux derivative $ f _ {G} ^ { \prime } ( x _ {0} ) $ of $ f $ at $ x _ {0} $. In other words, the Gâteaux gradient is defined by the formula

$$ f ( x _ {0} + h) = \ f ( x _ {0} ) + ( f _ {G} ^ { \prime } ( x _ {0} ), h) + \epsilon ( h), $$

where $ \epsilon ( th)/t \rightarrow 0 $ as $ t \rightarrow 0 $. In an $ n $- dimensional Euclidean space the Gâteaux gradient $ f _ {G} ^ { \prime } ( x _ {0} ) $ is the vector with coordinates

$$ \left ( \frac{\partial f ( x _ {0} ) }{\partial x _ {1} } \dots \frac{\partial f ( x _ {0} ) }{\partial x _ {n} } \right ) , $$

and is simply known as the gradient. The concept of the Gâteaux gradient may be extended to the case when $ X $ is a Riemannian manifold (finite-dimensional) or an infinite-dimensional Hilbert manifold and $ f $ is a smooth real function on $ X $. The growth of $ f $ in the direction of its Gâteaux gradient is larger than in any other direction passing through the point $ x _ {0} $.

How to Cite This Entry:
Gâteaux gradient. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=G%C3%A2teaux_gradient&oldid=47150
This article was adapted from an original article by V.M. Tikhomirov (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article