Generated by GPT-5-mini| Wilkinson, James H. | |
|---|---|
| Name | James H. Wilkinson |
| Birth date | 27 June 1919 |
| Death date | 5 October 1986 |
| Nationality | British |
| Occupation | Mathematician, numerical analyst |
| Known for | Numerical linear algebra, rounding error analysis, Wilkinson's polynomial |
Wilkinson, James H. was a British mathematician and pioneering numerical analyst whose work on numerical linear algebra, rounding error, and computational stability reshaped Numerical analysis and influenced practitioners at institutions such as National Physical Laboratory (United Kingdom), University of Manchester, and Harwell. He collaborated with contemporaries across the Royal Society, Courant Institute of Mathematical Sciences, and Princeton University networks, and his publications impacted software projects like EISPACK and influenced methods used at IBM and Argonne National Laboratory.
Born in Reigate, Wilkinson attended Dulwich College before studying at Gonville and Caius College, Cambridge, where he read mathematics under tutors associated with Trinity College, Cambridge influences and contemporaries linked to John von Neumann-era developments. His wartime service led him to work at Ministry of Supply establishments and to collaborations with researchers at Harwell and the National Physical Laboratory (United Kingdom), exposing him to problems later reflected in interactions with figures from Alan Turing's circles and the Royal Aircraft Establishment.
Wilkinson held positions at the National Physical Laboratory (United Kingdom), where he worked alongside scientists connected to Royal Society projects, and later at the University of Manchester, which had ties to the Manchester Mark 1 computing efforts. He spent sabbaticals and visiting appointments at institutions including the Courant Institute of Mathematical Sciences, Princeton University, and research centers collaborating with Argonne National Laboratory and IBM research groups. His career intersected with mathematicians from Stanford University, Massachusetts Institute of Technology, and the University of Cambridge who were advancing numerical computation and algorithmic design.
Wilkinson made foundational contributions to algorithms for solving eigenvalue problems related to the QR algorithm, and his analyses of rounding errors influenced implementations in libraries such as EISPACK and later LAPACK. He wrote seminal monographs and papers analyzing stability of matrix factorizations connected to LU decomposition and perturbation theory adjacent to work by Stewart, George W. and Golub, Gene H.. His study of ill-conditioned problems illuminated phenomena exemplified by the class of examples including Wilkinson's polynomial, which challenged assumptions held by numerical practitioners and prompted dialogue with theorists like Higham, Nicholas J. and Turing, Alan's successors. He advanced backward error analysis, a concept linked in practice to solution techniques used in packages at Argonne National Laboratory and influenced standards later codified in computational mathematics curricula at Princeton University and Cambridge. His published works engaged with eigenvalue sensitivity, matrix perturbation bounds, and iterative methods in conversation with developments originating from Von Neumann, John-inspired numerical frameworks and computational projects at Harwell.
Wilkinson received recognition from the Royal Society and institutions that confer prizes to leading scientists; his influence was acknowledged in memorial sessions hosted by organizations including SIAM and by fellowships connected to Cambridge University. He was invited to deliver lectures at venues such as Institute for Advanced Study and was cited in award notices from societies linked to Mathematical Association of America-type activities and to European academies collaborating with the Royal Society.
Wilkinson's legacy persists through citations in the literature of Numerical analysis, through algorithms embedded in software like EISPACK and LAPACK, and through the work of students and colleagues at institutions including University of Manchester, University of Cambridge, and Argonne National Laboratory. His name endures in educational settings at Gonville and Caius College, Cambridge and in lectures commemorating numerical pioneers such as Turing, Alan and Von Neumann, John. Contemporary texts by authors like Higham, Nicholas J. and Golub, Gene H. continue to reference his methodologies, ensuring his impact on computational practice at research centers including IBM and Courant Institute of Mathematical Sciences remains influential.
Category:British mathematicians Category:Numerical analysts Category:1919 births Category:1986 deaths