Leibniz cohomology for differentiable manifolds (Q1265652)

Ten years ago Loday proposed to study noncommutative Lie algebras, which he called Leibniz algebras. The associated Leibniz cohomology \(HL^*\) can be computed from the Koszul complex (computing Lie algebra cohomology) in which the exterior products are replaced by tensor products. Any Lie algebra \({\mathfrak g}\) is in particular a Leibniz algebra and the natural map from the tensor product to the exterior product induces a map \[ H^*_{\text{Lie}}({\mathfrak g},A)\to HL^*({\mathfrak g},A) \] for any \({\mathfrak g}\)-module \(A\). One of the main goals in the study of Leibniz algebras has been to link it to geometry. This paper is an interesting step in this direction. Let \(M\) be a smooth manifold, \({\mathfrak g}\) be the Lie algebra of its vector fields and \(A\) be the smooth functions on \(M\). Then the continuous Lie algebra cohomology of \({\mathfrak g}\) with coefficients in \(A\) is known as the Gelfand-Fuks cohomology. The author defines the Leibniz cohomology of \(M\) to be the corresponding continuous Leibniz cohomology of \({\mathfrak g}\) with coefficients in \(A\). One thus has a chain of natural maps \[ H^*_{\text{deR}}(M)\to H^*_{\text{GF}}(M)\to HL^*(M). \] \(HL^*(M)\) is a diffeomorphism invariant but not a homotopy invariant. Indeed, \(HL^*(\mathbb{R}^n)\) is non-trivial. Its computation in terms of universal invariants for foliations occupies the major part of the paper. Also of interest is the example of a Leibniz algebra that is not a Lie algebra which the author found in the work of Kac on vertex algebras.