TOPICS
Search

Stampacchia Theorem


"Stampacchia's theorem" is a name given to any number of related results in functional analysis, and while the body of the theorem often varies depending on the literature consulted, one commonly-encountered result attributed to Stampacchia is a sort of "representation inequality" for continuous, coercive bilinear forms on an arbitrary Hilbert space H. This particular version of the result is considerable for a number of reasons, most notably for its implication of the so-called Lax-Milgram theorem.

To state the above-mentioned version of the theorem, let H be a Hilbert space, let a be a continuous and coercive bilinear form on H, and let K be a closed and convex subset of H. One result of Stampacchia says that, under these assumptions, any function f in H necessarily corresponds to a unique function u in K for which the inequality

 a(u,v-u)>=<f,v-u>_H
(1)

is satisfied for all functions v in K where here, <·,·>_H denotes the inner product on H. Note that this form of the result is particularly nice in the sense that, by examining an arbitrary element f in H and by selecting an element u in K for which the above inequality holds for all v in V, the convexity of K implies that

 a(u,v)>=<f,v>_H
(2)

for all v in H, and because of the bilinearity of both a and of the inner product, the above inequality restated for -v necessarily yields

 a(u,v)>=<f,v>_H.
(3)

Combining these latter two inequalities and considering the case for K=H, one then arrives at the Lax-Milgram theorem which states that, under the assumptions stated above, any element f in H necessarily corresponds to a unique element u in H satisfying

 a(u,v)=<f,v>_H
(4)

for all v in H.

As noted above, however, the content of the results attributed to Stampacchia may be different. Yet another common form of the theorem states that if a function u lies in the Sobolev space W_0^(1,p)(Omega) for a bounded domain Omega subset R^n and if G:R->R is a real-valued Lipschitz function satisfying G(0)=0, then the composition G(u) also lies in W_0^(1,p)(Omega) provided that G(u) in L^p(Omega) and that the function G satisfies the generalized derivative identity

 del G(u)=G^'(u)del u
(5)

almost everywhere in Omega. In the above, W_0^(1,p)(Omega) is defined to be the collection of functions in W^(1,p)(Omega) with zero trace, i.e.,

 W_0^(1,p)(Omega)={u in W^(1,p)(Omega):there exists {u_m}_(m=1)^infty subset C_c^infty(Omega) such that u_m->u in W^(1,p)(Omega)}
(6)

where C_c^infty(Omega) is the collection of all smooth functions on Omega with compact support.


See also

Lax-Milgram Theorem

This entry contributed by Christopher Stover

Explore with Wolfram|Alpha

References

Dimitrios, K. "The Stampacchia and Lax-Milgram Theorems and Applications." http://www.stat-athens.aueb.gr/gr/master/sumschool/files/Kravvaritis.pdf.Monteillet, A. "A Theorem of Stampacchia." http://aurelien.monteillet.com/Stages/Stampacchia-anglais.pdf.Stampacchia, G. "Équations elliptiques du second ordre à coefficients discontinus." Séminaire Jean Leray 3, 1-77, 1963-1964.

Cite this as:

Stover, Christopher. "Stampacchia Theorem." From MathWorld--A Wolfram Web Resource, created by Eric W. Weisstein. https://mathworld.wolfram.com/StampacchiaTheorem.html

Subject classifications