作者: Madeleine Udell , Karanveer Mohan , David Zeng , Jenny Hong , Steven Diamond
关键词: Proper convex function 、 Mathematics 、 Subderivative 、 Convex optimization 、 Linear matrix inequality 、 Mathematical optimization 、 Convex analysis 、 Conic optimization 、 Convex hull 、 Algorithm 、 Absolutely convex set
摘要: This paper describes Convex1, a convex optimization modeling framework in Julia. Convex translates problems from user-friendly functional language into an abstract syntax tree describing the problem. concise representation of global structure problem allows to infer whether complies with rules disciplined programming (DCP), and pass suitable solver. These operations are carried out Julia us- ing multiple dispatch, which dramatically reduces time required verify DCP compliance parse conic form. then automatically chooses appropriate backend solver solve form