This is another part of myself as a scientist.
I'm using and developing the 半分-限定された programming solver, and the SDPA-GMP has been 申し込む／申し出d by the NEOS server. It's 広大な／多数の／重要な 栄誉（を受ける） to me! because it was recommended by others, not by our team.
I'm a 地位，任命する doc at RIKEN. Not many know about RIKEN, so I'd like to explain a bit. RIKEN was 設立するd in 1917 and famous 研究室／実験室 in Japan. Prof. Nishina, Prof. Yukawa and Prof. Tomonaga were also there. We have very large 粒子 accelerator before the world war II, 行為／行うd by Prof. Nishina. Prof. Yukawa and Tomonaga were awarded Nobel prize (physics).
My 研究 利益／興味 is the 濃度／密度 matrix theory, and my 動機づけ is solving Schroedinger equation, which is the 根底となる equation of Chemistry. This is a 部分的な／不平等な differential equation of the second order. It is known that we can 明確に表す as Semidefinite programming, and we solved 正確に／まさに for the first time!.
It is very difficult to solve the semidefinite programming exacltly and efficiently, and lot of unknowns are there 予定 to inaccuracy of the results.
In 2006 or so I started to develop the SDPA-GMP. Some problems of semidefinite programing are very difficult 予定 to inaccuracy of "二塁打". We need more 重要な digits, say 100, where 二塁打 has only 16. GMP is one of the fastest 実施 of 多重の precision arithmetic library, and I use it to 器具／実施する some BLAS/LAPACK codes. This is almost dumb
translation of BLAS/LAPACK code to the GMP, and very slow (usually
500 times slower than 二塁打). There are two papers 関係のある to SDPA-GMP...Variational 計算／見積り of second-order 減ずるd 濃度／密度 matrices by strong N-representability 条件s and an 正確な semidefinite programming solver
and Strange 行為s of 内部の-point Methods for Solving Semidefinite Programming Problems in Polynomial Optimization (preprint)
I'm using and developing the 半分-限定された programming solver, and the SDPA-GMP has been 申し込む／申し出d by the NEOS server. It's 広大な／多数の／重要な 栄誉（を受ける） to me! because it was recommended by others, not by our team.
I'm a 地位，任命する doc at RIKEN. Not many know about RIKEN, so I'd like to explain a bit. RIKEN was 設立するd in 1917 and famous 研究室／実験室 in Japan. Prof. Nishina, Prof. Yukawa and Prof. Tomonaga were also there. We have very large 粒子 accelerator before the world war II, 行為／行うd by Prof. Nishina. Prof. Yukawa and Tomonaga were awarded Nobel prize (physics).
My 研究 利益／興味 is the 濃度／密度 matrix theory, and my 動機づけ is solving Schroedinger equation, which is the 根底となる equation of Chemistry. This is a 部分的な／不平等な differential equation of the second order. It is known that we can 明確に表す as Semidefinite programming, and we solved 正確に／まさに for the first time!.
It is very difficult to solve the semidefinite programming exacltly and efficiently, and lot of unknowns are there 予定 to inaccuracy of the results.
In 2006 or so I started to develop the SDPA-GMP. Some problems of semidefinite programing are very difficult 予定 to inaccuracy of "二塁打". We need more 重要な digits, say 100, where 二塁打 has only 16. GMP is one of the fastest 実施 of 多重の precision arithmetic library, and I use it to 器具／実施する some BLAS/LAPACK codes. This is almost dumb
translation of BLAS/LAPACK code to the GMP, and very slow (usually
500 times slower than 二塁打). There are two papers 関係のある to SDPA-GMP...Variational 計算／見積り of second-order 減ずるd 濃度／密度 matrices by strong N-representability 条件s and an 正確な semidefinite programming solver
and Strange 行為s of 内部の-point Methods for Solving Semidefinite Programming Problems in Polynomial Optimization (preprint)
