Calculus of Variations: The Gateaux vs First Variation Debate

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
1 reply · 6K views
rdt2
Messages
124
Reaction score
2
Older textbooks on the Calculus of Variations seem to define the first variation of a functional [tex]\Pi[/tex] as:

[tex]\delta \Pi = \Pi(f + \delta f) - \Pi (f)[/tex]

which looks analogous to:

[tex]\delta f = \frac {df} {dx} \delta x = lim_{\delta x \rightarrow 0} (f(x+ \delta x) -f(x))[/tex]

from differential calculus. However, newer books seem to define the first variation as the Gateaux derivative:

[tex]\left[ \frac {d} {d \epsilon} \Pi (f+ \epsilon h) \right]_{\epsilon = 0 }[/tex]

which looks more like the gradient [tex]\frac {df} {dx}[/tex] than the difference [tex]\delta x[/tex]. Which is the better 'basic' definition?
 
Physics news on Phys.org
The first definition is similar to the discrete [itex]\Delta[/itex] operator in real analysis. The 2nd def. is similar to the continuous d operator.