The Borsuk-Ulam type theorems for finite-dimensional compact group actions (Q2169240)

This paper is about the existence of equivariant maps between \(G\)-spaces. The result extends several known results, notably a result by \textit{M. Clapp} and \textit{D. Puppe} [Differ. Geom. Appl. 6, No. 4, 367--396 (1996; Zbl 0874.53035)] and a result by \textit{C. Biasi} and \textit{D. de Mattos} [Bull. Braz. Math. Soc. (N.S.) 37, No. 1, 127--137 (2006; Zbl 1102.55001)]. Many of the known related results are for actions of Lie Groups, but here the group \(G\) acting is more general. The main result of the paper under review is as follows: Theorem 3.6 Let \(G\) be a finite-dimensional compact group, and let \(X\), \(Y\) be path connected, paracompact free \(G\)-spaces. Suppose that for some natural \(m \geq 1\), \(H^i(X;k) = \{0\}\) for \(0 < i < m\) and \(H^{m+1} (Y/G;k) = \{0\}\). Then, if \(\beta_m(X;k) < \beta_m(B_G;k)\), there is no \(G\)-equivariant map \(f :X\to Y\). Also a result is provided for the case where the group \(G\) is a finite-dimensional pro-torus, extending the result by Clapp and Puppe [loc. cit.]. In order to overcome the difficulty which arises from the fact that \(G\) is not a compact Lie Group, as pointed out in the introduction, ``it is used the fact that there is a closed normal subgroup \(N\) in a finite-dimensional compact group \(G\), such that \(G/N\) is a compact Lie group and then applying the cohomological technique of Lie group actions to show the main theorems''. Nevertheless the proofs are not completely straightforward.