Results of the EMO'2017 Real-world Problems Track
"Raw" Result Data
On each problem participants were judged by the overall dominated hypervolume within the given budget of function evaluations. There were 10 participants in the field. The best function value per problem and participant (10 times 10 double precision numbers) is listed in this text file.
Participants were ranked based on aggregated problem-wise ranks (details here and here). The following results table lists participants with overall scores (higher is better) and the sum of ranks over all problems (lower is better) The table can be sorted w.r.t. these criteria (however, for this track the rankings happen to coincide).
|rank||participant||method name||method description||software||paper||score||sum of ranks|
|1||Simon Wessing||Model-based HV-maximization||
3n points initial sample, 12 points alternating single-objective EI, then model-based HV-maximization until 250 evals are consumed. If Budget <= 500 restart, else (50+5)-SMS-EMOA and model-based refinement (again HV-maximization) with the last 100 evals.
|2||Julien Bect||BMOO||EHVI / GP / SMC||link||8.15735||32|
|4||John T||Modified multi-objective mutation-based||5.15331||43|
|5||Daniel Horn||SMS-EGO||SMS-EGO implementation from R-package mlrMBO||link||link||4.66084||47|
|6||DIKU, University of Copenhagen||2.63764||60|
|7||Al Jimenez||Curved Trajectories Algorithm (CTA)||
Curved trajectories algorithm (CTA) is a package for the minimization of unconstrained functions of several variables with intervals on the variables. The core algorithm is novel in that steps may follow polynomial space curves instead of straight lines. The space curves result from truncations of a Taylor series expansion of the Gradient inverse function. A critical item in the efficiency of CTA is the factorization of the sparse Hessian matrix and handling a non-positive definite Hessian. This paper describes a new approach for non-positive definite Hessians that has given robust results especially for ill-conditioned problems and the CTA package compares very favorably with other minimization packages using a sample of large Cuter problems.
|Contact author firstname.lastname@example.org||link||1.0199||68|
|8||Artelys||Artelys Knitro||Artelys Knitro used in derivative-free mode with multistart||link||link||1.0116||75|