Techniques, computations, and conjectures for semi-topological \(K\)-theory (Q707437)

Let \(X\) be a complex variety and let \(K^{\text{sst}}(X)\) be the singular semi-topological \(K\)-theory of \(X\) as defined by two of the authors [see for example \textit{E. Friedlander} and \textit{M. Walker}, ``Semi-topological \(K\)-theory using function complexes'', Topology 41, 591--644 (2002; Zbl 1003.19003)]. There are natural maps \[ K_{q}(X)\to K^{\text{sst}}(X)\to ku^{-q}(X^{\text{an}}), \quad q\geq 0, \] where \(K_{q}(X)\) is the usual \(K\)-theory of \(X\), and \(ku^{-q}(X^{\text{an}})\) denotes the complex topological \(K\)-theory of the underlying analytic space of \(X\). Let \(X\) now be a smooth quasi-projective complex variety, in this case the authors construct a spectral sequence that strongly converges to \(K_{*}^{\text{sst}}(X)\) where the \(E_{2}\)-term is given by \(E_{2}^{p,q}\cong L^{-q}H^{p-q}(X)\), where \(L^{*}H^{*}(X)\) denotes the morphic cohomology of \(X\). This spectral sequence is so natural that maps to the Atiyah-Hirzebruch spectral sequence that computes \(ku(X)\), even more is true: Let \(X\) be a smooth, quasi-projective variety, and \(A\) be any abelian group. Then, there are maps of strongly convergent spectral sequences \[ E^{p,q}_{2}(\text{alg})\to E^{p,q}_{2}(\text{sst})\to E^{p,q}_{2}(\text{top}), \] where the first term is the motivic cohomology spectral sequence that computes algebraic \(K\)-theory with \(A\) coefficients, the mid-term is the one constructed in this paper and computes \(K^{\text{sst}}(X;A)\), and the last term is the Atiyah-Hirzebruch spectral sequence computing complex topological \(K\)-theory with coefficients in \(A\). Furthermore, these maps of spectral sequences induce the maps mentioned above in the abutments. Another part of the paper is devoted to study the term \(L^{*}H^{*}(X)\), the authors use that for smooth varieties, this term is isomorphic to the dual of Lawson homology to obtain a factorization of the generalized cycle map from Lawson homology to Borel-Moore homology through an object discovered by the authors. Finally, as an application, the authors prove that the natural map \(K_{n}^{\text{sst}}(X)\to ku^{-n}(X^{\text{an}})\) is an isomorphism for \(n\geq \dim(X)-1\) and a monomorphism for \(n=\dim(X)-2\) for a large class of complex varieties that includes, for example, smooth quasi-projective toric varieties.