The proof of the \(l^2\) decoupling conjecture (Q2352913)

The main result can be formulated as follows. Let \(N_\delta\) be the \(\delta\)-neighborhood of a compact \(C^2\) hypersurface in \(\mathbb R^n\), \(n\geq2\), with positive definite second fundamental form, and let \(P_\delta\) be a finitely overlapping cover of it with special curved regions \(\theta\), each sitting inside a \(\delta^{1/2}\times...\times\delta^{1/2} \times\delta\) rectangular box (approximately). The Fourier restriction of \(f\) to \(\theta\in P_\delta\) will be denoted by \(f_\theta\). If the support of the Fourier transform of \(f\) is within \(N_\delta\), then for \(p\geq\frac{2(n+1)}{n-1}\) and \(\varepsilon>0\), \[ \|f\|_p\leq C_\varepsilon\delta^{-\frac{n-1}4+\frac{n+1}{2p}-\varepsilon}\left(\sum\limits_\theta \|f_\theta\|_p^2\right)^{1/2}. \] This is a serious improvement of earlier results. It has many interesting consequences, some of them are discussed in the paper.