Four-dimensional wall-crossing via three-dimensional field theory (Q5962381): Difference between revisions

The main topic of this paper is a wall-crossing formula (WCF) for the degeneracies of BSP states in quantum field theories with \(d=4\), \({\mathcal N}=2\) supersymmetry. The only available index for \(d=4\), \({\mathcal N}=2\) supersymmetry is the second helicity supertrace denoted by \(\Omega(\gamma; u)\) which plays an important role in the Seiberg-Witten theory. \textit{M. Kontsevich} and \textit{Y. Soibelman} [in: ``Stability structures, motiving Donaldson-Thomas invariants and cluster transformations'' (2008), \url{http://arXiv.org/abs/0811.2435v1[math.AG]}] have proposed a remarkable wall-crossing formula (denoted by KS formula) for the \(\Delta{\Omega}\) which applies to all possible decays. In this paper the authors offer a physical interpretation and proof of the KS formula in the case of \(d=4\), \({\mathcal N}=2\) field theories. They also give an exact description of the BPS instanton corrections to the hyper-Kähler metric of the moduli space of the theory on \(\mathbb R^3\times S^1\), where \(S^1\) has radius \(R\). Indeed, they consider the gauge theory on the space \(\mathbb R^3\times S^1\). At low energies this theory is described by a \(d=3\) sigma model with hyper-Kähler target space \(({\mathcal M},g)\). This sigma model receives corrections from BPS instantons, in which the world-line of a BPS particle of the \(d=4\) theory is wrapped around \(S^1\). Expanding the metric \(g\) at large \(R\), one can therefore read off the degeneracies \(\Omega(\gamma;u)\) of the BPS particles. We know that the \(\Omega(\gamma;u)\) are discontinuous, but \(g\) must be continuous. This is, in effect, based on the physical principal that the only singularities in the low-energy effective field theory Lagrangian arise from the appearance at special moduli of massless particles. Therefore the wall-crossing formula reduces to the statement that this metric is continuous. In their construction of the metric, they use a four-dimensional analogue of the two-dimensional \(tt^*\) equations.