Convergence Guarantees

Cross-source consensus on Convergence Guarantees from 1 sources and 5 claims.

1 sources · 5 claims

How it works

The convergence theorem requires convexity, a saddle point, solvable subproblems, bounded parameters, and summable parameter changes. — Learning Over-Relaxation Policies for ADMM with Convergence Guarantees
The theorem guarantees convergence of primal residuals, dual residuals, and objective values. — Learning Over-Relaxation Policies for ADMM with Convergence Guarantees
The theorem does not guarantee convergence to a unique primal-dual solution. — Learning Over-Relaxation Policies for ADMM with Convergence Guarantees
The convergence argument controls learned adaptation by constraining parameter sequences instead of relying on training itself. — Learning Over-Relaxation Policies for ADMM with Convergence Guarantees
The summable-change condition is experimentally enforced for Gamma_k by stopping updates after 500 OSQP iterations. — Learning Over-Relaxation Policies for ADMM with Convergence Guarantees