Genetic Improvement of Data for Maths Functions

We use continuous optimisation and manual code changes to evolve up to 1024 Newton-Raphson numerical values embedded in an open source GNU C library glibc square root sqrt to implement a double precision cube root routine cbrt, binary logarithm log2 and reciprocal square root function for C in seconds. The GI inverted square root x -1/2 is far more accurate than Quake’s InvSqrt, Quare root. GI shows potential for automatically creating mobile or low resource mote smart dust bespoke custom mathematical libraries with new functionality.

The Rectangle Covering Number of Random Boolean Matrices

The rectangle covering number of an $n$-by-$n$ Boolean matrix $M$ is the smallest number of 1-rectangles which are needed to cover all the 1-entries of $M$. Its binary logarithm is the Nondeterministic Communication Complexity, and it equals the chromatic number of a graph $G(M)$ obtained from $M$ by a construction of Lovasz and Saks.We determine the rectangle covering number and related parameters (clique size, independence ratio, fractional chromatic number of $G(M)$) of random Boolean matrices, where each entry is 1 with probability $p = p(n)$, and the entries are independent.

Decomposable graphs and definitions with no quantifier alternation

International audience Let $D(G)$ be the minimum quantifier depth of a first order sentence $\Phi$ that defines a graph $G$ up to isomorphism in terms of the adjacency and the equality relations. Let $D_0(G)$ be a variant of $D(G)$ where we do not allow quantifier alternations in $\Phi$. Using large graphs decomposable in complement-connected components by a short sequence of serial and parallel decompositions, we show examples of $G$ on $n$ vertices with $D_0(G) \leq 2 \log^{\ast}n+O(1)$. On the other hand, we prove a lower bound $D_0(G) \geq \log^{\ast}n-\log^{\ast}\log^{\ast}n-O(1)$ for all $G$. Here $\log^{\ast}n$ is equal to the minimum number of iterations of the binary logarithm needed to bring $n$ below $1$.

An estimate of mean efficiency of search trees for arbitrary sets of binary words

We consider the problem on constructing a binary search tree for an arbitrary set of binary words, which has found a wide use in informatics, biology, mineralogy, and other fields. It is known that the problem on constructing the tree of minimal cost is NP-complete; hence the problem arises to find simple algorithms which allow us to construct trees close to the optimal ones. In this paper we demonstrate that even simplest algorithm yields search trees which are close to the optimal ones in average, and prove that the mean number of nodes checked in the optimal tree differs from the natural lower bound, the binary logarithm of the number of words, by no more than 1.04.