Resolution of Peller's problem concerning Koplienko-Neidhardt trace formulae (Q2827964)

The main result of the article under review is a negative solution to the problem posed in [\textit{B. S. Pavlov}, Probl. Mat. Anal. 2, 99--122 (1969; Zbl 0211.17201)] on the validity of the Koplienko-Neidhardt trace formulae. Namely, the authors prove that there exist a \(C^2\) function \(f\) on the real line with bounded second derivative, a self-adjoint operator \(A\) and a Hilbert-Schmidt self-adjoint operator \(B\) such that the operator NEWLINE\[NEWLINE f(A+B)-f(A)-\frac{d}{dt}(A+tB)\Big|_{t=0} NEWLINE\]NEWLINE does not belong to the trace class.