Answers are grounded in this paper and cite their sources.
Letter reports on BIBS stability in in-parameter-linear neural architectures, suggesting implications for gradient learning algorithms.
This letter summarizes and proves the concept of bounded-input bounded-state (BIBS) stability for weight convergence of a broad family of in-parameter-linear nonlinear neural architectures (IPLNAs) as it generally applies to a broad family of incremental gradient learning algorithms. A practical BIBS convergence condition results from the derived proofs for every individual learning point or batches for real-time applications.
No takes yet. Share an insight, caveat, or question.
Bukovský et al. (2021) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: