Math 2.0

Potential theory Definition

Dirichlet form

Builds on MeasureHilbert spaceLp spaceBilinear formSobolev spaceSelf-adjoint operator

Definition

Let (X,𝒜,m)(X, \mathcal{A}, m) be a σ\sigma-finite measure space. Let H=L2(X,m)H = L^2(X, m) be the real Hilbert space of square-integrable functions (an Lp space), with inner product (⋅,⋅)(\cdot, \cdot). Let ℱ\mathcal{F} be a dense linear subspace of HH and let ℰ ⁣:ℱ×ℱ→ℝ\mathcal{E} \colon \mathcal{F} \times \mathcal{F} \to \R be a bilinear form. The pair (ℰ,ℱ)(\mathcal{E}, \mathcal{F}) is a Dirichlet form on L2(X,m)L^2(X, m) if:

  1. ℰ\mathcal{E} is symmetric: ℰ(u,v)=ℰ(v,u)\mathcal{E}(u, v) = \mathcal{E}(v, u) for all u,v∈ℱu, v \in \mathcal{F}.
  2. ℰ\mathcal{E} is nonnegative: ℰ(u,u)≥0\mathcal{E}(u, u) \ge 0 for all u∈ℱu \in \mathcal{F}.
  3. ℰ\mathcal{E} is closed: ℱ\mathcal{F} is complete for the norm ∥u∥1=(ℰ(u,u)+(u,u))1/2\norm{u}_1 = \left(\mathcal{E}(u, u) + (u, u)\right)^{1/2}.
  4. ℰ\mathcal{E} is Markovian: for every u∈ℱu \in \mathcal{F}, the function v=min⁡(max⁡(u,0),1)v = \min(\max(u, 0), 1) lies in ℱ\mathcal{F} and ℰ(v,v)≤ℰ(u,u)\mathcal{E}(v, v) \le \mathcal{E}(u, u).

Remarks

Example: on L2(ℝn)L^2(\R^n) with Lebesgue measure, take ℱ\mathcal{F} to be the Sobolev space W1,2(ℝn)W^{1,2}(\R^n) and ℰ(u,v)=∫ℝn∇u⋅∇v dx\mathcal{E}(u, v) = \int_{\R^n} \nabla u \cdot \nabla v \, dx. It is symmetric and nonnegative, and ∥⋅∥1\norm{\cdot}_1 is the Sobolev norm, which is complete. The truncation v=min⁡(max⁡(u,0),1)v = \min(\max(u, 0), 1) of u∈W1,2u \in W^{1,2} lies in W1,2W^{1,2}, with ∇v=∇u\nabla v = \nabla u almost everywhere on {0<u<1}\set{0 < u < 1} and ∇v=0\nabla v = 0 almost everywhere elsewhere, so ℰ(v,v)≤ℰ(u,u)\mathcal{E}(v, v) \le \mathcal{E}(u, u).

Non-example: on L2(ℝ)L^2(\R) take ℱ=W2,2(ℝ)\mathcal{F} = W^{2,2}(\R) and ℰ(u,v)=∫u′′v′′ dx\mathcal{E}(u, v) = \int u'' v'' \, dx. Clauses 1 to 3 hold, but clause 4 fails. If u∈W2,2(ℝ)u \in W^{2,2}(\R) is smooth and crosses the level 11 with nonzero slope, its truncation has a corner there, so it is not in W2,2(ℝ)W^{2,2}(\R).

A closed, symmetric, nonnegative, densely defined form corresponds to a nonnegative self-adjoint operator AA with ℰ(u,v)=(A1/2u,A1/2v)\mathcal{E}(u, v) = (A^{1/2} u, A^{1/2} v) and ℱ\mathcal{F} the domain of A1/2A^{1/2}. Clause 4 holds if and only if the semigroup Tt=e−tAT_t = e^{-tA} is Markovian: 0≤f≤10 \le f \le 1 almost everywhere implies 0≤Ttf≤10 \le T_t f \le 1 almost everywhere, for every t>0t > 0. Let XX be a locally compact separable metric space and mm a Radon measure of full support. The form is then regular if ℱ∩Cc(X)\mathcal{F} \cap C_c(X) is dense in ℱ\mathcal{F} for ∥⋅∥1\norm{\cdot}_1 and dense in Cc(X)C_c(X) for the uniform norm.

Reading path

To understand this page from the ground up, read these 12 pages in order. The foundations come first.

Show the path as a graph
Each page uses the pages above it. The colour of the bar on a box shows how strong the evidence is, and the box also gives it as text. The list has the same pages.
  • no evidence level
  1. Bilinear form concept
  2. Countable additivity concept
  3. Dirac measure concept
  4. Extended real number line object
  5. Hilbert space concept
  6. Lp space concept
  7. Measurable space concept
  8. Self-adjoint operator concept
  9. Series concept
  10. Sobolev space concept
  11. Measure definition
  12. Dirichlet form definition

Further reading

  1. Web https://en.wikipedia.org/wiki/Dirichlet_form
About this page
TagDW5G0SQ2G5
URIm2://DW5G0SQ2G5
Kinddefinition
WritingW0 stub
MSC31C25, 60J46
DefinesDirichlet form
UsesMeasure, Hilbert space, Lp space, Bilinear form, Sobolev space, Self-adjoint operator

Markdown · JSON · Edit on GitHub

What this page needs for W1

  • an inventory row: the row of the inventory that the node accounts for (0 of 1)
  • a source with a locator (0 of 1)

The writing levels say how complete the text of a page is. They do not measure its truth.

What links here

No page links to this page yet.