Answers are grounded in this paper and cite their sources.
A major drawback of recursive adaptive filters based on gradient methods is that convergence to a global minimum is not always achieved. This is due to a nonconvex mean square error (MSE) performance surface. This article develops a continuous-time least mean square algorithm that converges to the global minimum with probability one.
No takes yet. Share an insight, caveat, or question.
Edmonson et al. (1999) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: