Fredholm theory and transversality for the parametrized and for the \(S^{1}\)-invariant symplectic action (Q990222)

Hamiltonian Floer homology is commonly referred to as Morse homology for the symplectic action functional on the free loop space of a symplectic manifold. The free loop space carries an \(S^1\)-action. It was realized at an early stage of the theory that Floer homology should admit an \(S^1\)-equivariant version. Such an \(S^1\)-equivariant theory was first defined by Viterbo in the context of symplectic homology. The article under review grew out of efforts to understand \(S^1\)-equivariant Floer homology and the desire to put it on firm ground. The authors' initial motivation was the desire to interpret their long exact sequence for contact and symplectic homology as a Gysin exact sequence. The authors study the parametrised Hamiltonian action functional for finite-dimensional families of Hamiltonians. They prove that the linearized operator for the \(L^2\)- gradient lines is Fredholm and surjective for a generic choice of Hamiltonian and almost complex structure. The authors also establish the Fredholm property and transversality for generic \(S^1\)-invariant families of Hamiltonians and almost complex structures, parametrized by odd-dimensional spheres. Then, they define \(S^1\)-equivariant Floer homology. The results of the present paper will be instrumental in proving that transversality can be achieved for linearized contact homology, modulo the data of a linearization for the contact complex. The author's methods can be extended to define parametrized Floer homology groups for a symplectic fibration. As an intermediate result of independent interest, the authors generalize Aronszajin's unique continuation theorem to a class of elliptic integro-differential inequalities of order two.