Scalar and asymptotic scalar derivatives. Theory and applications (Q2469761)

The scalar derivative \(f^\#(x_0)\) of a mapping \(f:\mathbb R^n\to \mathbb R^n\) at a point \(x_0\in \mathbb R^n\) is defined by the condition \(f^\#(x_0)=\lim_{x\to x_0}\langle f(x)-f(x_0),x-x_0\rangle/\|x-x_0\|^2, \) provided this limit exists and is finite, a notion introduced and studied by the second author of this book. It turns out that this notion is related to skew linear operators on \(\mathbb R^n\); a linear operator \(A:\mathbb R^n\to \mathbb R^n\) is scalarly differentiable iff it is of the form \(A=B+cI_n,\) where \(B\) is a skew linear operator on \(\mathbb R^n.\) In the case \(n=2,\) a mapping \(f: \mathbb C\to \mathbb C\) is \(\mathbb C\)-differentiable at \(z_0\) iff it is both \(\mathbb R\)-differentiable and scalarly differentiable at \(z_0.\) Taking \(\liminf\) or \(\limsup\) in the above definition, one obtains the notions of lower scalar derivative and upper scalar derivative, respectively. The basic properties of scalar derivatives are studied in the first chapter, ``Scalar derivatives in Euclidean spaces'', emphasizing their relations with directional derivatives, monotonicity properties, etc. A mapping \(f:E\to F\), \(E,F\) Banach spaces, is called asymptotically linear if there exists a continuous linear operator \(L:E\to F\) such that \(\lim_{\|x\|\to \infty}\|f(x)-L(x)\|/\|x\|=0.\) In this case, the operator \(L\) is unique and is called the asymptotic derivative of \(f,\) denoted by \(f_\infty.\) This notion was considered first by M. A. Krasnoselskij who called a mapping \(f\) satisfying a condition of this kind asymptotically linear. The second chapter of the book, ``Asymptotic derivatives and asymptotic scalar derivatives'', is devoted to the study of asymptotic derivatives, including a discussion of their relations with the stability of the Hyers-Ulam property and fixed point properties. Extensions to locally convex spaces are considered, too. This chapter contains also a study of asymptotic scalar derivatives, defined by the first author of the book via a semi-inner product \([\cdot,\cdot]\) (in the sense of Lumer) by the condition \(\lim_{\|x\|\to \infty}(f(x)-[T(x),x])/[x,x]=0. \) In Chapter 3, ``Scalar derivatives in Hilbert spaces'', the notion of scalar derivative is studied within the framework of Hilbert spaces, emphasizing its strong connection with submonotone mappings as considered by J. E. Spingarn (1982) and P. Georgiev (1997). Applications to fixed point theorems of Krasnoselskij type, to duality in complementarity and to the inversion of set-valued mappings are included. In Chapter 4, ``Scalar derivatives in Banach spaces'', the method of scalar derivatives is applied to obtain generalizations of some fixed point theorems (by Altman and Krasnoselskij) involving various measures of noncompactness. In the last chapter of the book, Chapter 5, ``Monotone vector fields on Riemannian manifolds'', scalar derivatives are studied on Riemannian manifolds, with emphasis on Hadamard manifolds, meaning complete, simply connected Riemannian manifolds of nonpositive sectional curvature. The basic notions are those of geodesic monotone vector fields and geodesic scalar derivatives, which are natural extensions to this framework of those considered in the first chapter (the case \(\mathbb R^n\)). Based mainly on recent results obtained by the authors, appearing for the first time in book form, this volume is a valuable contribution to the field. It can be recommended to researchers and graduate students working in nonlinear analysis, fixed point theory, optimization, Riemannian geometry, and applied mathematics.