3
$\begingroup$

Let $(X, \omega)$ be a symplectic manifold. For any non-empty subset $Y \subset X$ we may define the displacement energy as $$ e(Y)=\mathrm{inf}\{||\phi||_H \: | \phi \in Ham(X, \omega), \phi(Y) \cap Y=\emptyset\} $$ where $||\cdot||$ is the Hofer norm.

It and its extension to symplectomorphisms isotopic to identity have been studied in literature (Banyaga, Eliashberg-Polterovich, McDuff-Lalonde). For instance, it is known that for any non-empty open subset energy is strictly positive.

My question is: was an analogue of the displacement energy for several intersection points, i.e. $$ e_k(Y)=\mathrm{inf}\{||\phi||_H \: | \phi \in Ham(X, \omega), |\phi(Y) \cap Y|\leqslant k\}, $$ ever studied?

It is of course necessary to make sense of the 'number of intersection points'; if we restrict to smooth submanifolds I think there is well-defined notion of intersection multiplicity.

My interest in this modified energy is that in some situations it can see more symplectic geometry than its traditional counterpart. For instance, for suitably chosen Lagrangian submanifold with non-trivial Floer cohomology $L \subset X^{2n}$ the usual energy $e(L)=e_0(L)$ is $+\infty$ while $e_k(L)$ with $k=\sum_{i=0}^{n}b_i(L; \mathbb{Z}_2)$ is finite (the choice of $k$ is explained, of course, by Arnold conjecture).

$\endgroup$
1
  • 1
    $\begingroup$ I'm not sure if this is along the lines you have in mind, but Biran and Cornea discuss a conjectural dichotomy between "wide" and "narrow" Lagrangians. Their conjecture, if I understood correctly, is that the Floer homology of L with itself, when defined, is either 0 or is the homology of L (where this is dependent on the choice of coefficients). If this is correct (and I don't know the state of the conjecture), then it would give a rough indication that $e_0$ and $e_k$ are the interesting ones in your family. $\endgroup$
    – Sam Lisi
    Commented Mar 23, 2018 at 14:46

0

You must log in to answer this question.