JuMP-dev 2018 | Developing new optimization methods with JuliaSmoothOptimizers | Abel Siqueira

We present the packages in JuliaSmoothOptimizers (JSO). Aimed at nonlinear optimization in Julia, we develop tools to help create, test, and compare optimization solvers. Among other things, we have trust-region and line-search methods for subproblems, well-known collections of problems, like the CUTEst collection (Gould et al., 2014), the ability to easily create new problems, and tools for benchmarking the solvers, such as the performance profile. We show that, while sharing many common goals with other optimization packages and groups, JSO works on an underexplored area, and integrates well with other packages - specially JuMP.

00:00 Welcome!
00:10 Help us add time stamps or captions to this video! See the description for details.