ResearchBenchmarksNetworked control problem on Opinion dynamics De Groot model N=10 agents 80,000 iterations (train)Follow75.63Control Objective ScoreIDO73.23873.85974.4875.101Mar 24, 2026Evaluation ResultsMethodMethodLinksControl Objective ScoreIDOAlgorithm/Loss=Relativ...Algorithm/Loss=Relative Entropy2026.0375.63IDOAlgorithm/Loss=Adjoint...Algorithm/Loss=Adjoint Matching2026.0374.69IDOAlgorithm/Loss=Log-var...Algorithm/Loss=Log-variance loss2026.0374.52EIGFAlgorithm/Loss=Relativ...Algorithm/Loss=Relative loss2026.0373.33