A conjectural improvement for inequalities related to regulators of number fields (Q2072248)

Following Remak's remark in [\textit{R. Remak}, Compos. Math. 10, 245--285 (1952; Zbl 0047.27202)] that number fields with small regulators have small discriminants, and Friedman's generalization in [\textit{E. Friedman}, Invent. Math. 98, No. 3, 599--622 (1989; Zbl 0694.12006)], one can search all the number fields with a fixed signature and whose regulator is less than a prescribed bound. Exhaustive and partial results about this problem can be find in [\textit{S. Astudillo} et al., J. Number Theory 167, 232--258 (2016; Zbl 1415.11158)], where the authors give a list of fields with minimal or small regulators for degrees less than 7 -- except for the signature \((5,1)\) -- and a partial list for degrees 8 and 9. This paper completes those previous results in degree 8, and refines them in degree 5 and 7, following Friedman and Ramirez-Raposo's ideas found in [\textit{E. Friedman} and \textit{G. Ramirez-Raposo}, J. Number Theory 198, 381--385 (2019; Zbl 1429.11200)], where the upper bound in Remak-Friedman's inequality have been improved, allowing the two authors to solve the missing \((5,1)\)-signature case in degree 7. However, the list of the fields with small regulators given in the paper under review is based on conjectural upper bounds, the validity of those conjectures being extensively discussed by the author of the present paper, and corroborated by numerical experiments conducted on/with PARI/GP and the MATLAB Optimization Toolbox.