# kaal:claim:4796714-031

**Claim.** Decentralized Federated Learning reaches the global minimum with zero performance gap and matches the convergence rate of centralized methods when the loss function is smooth and strongly convex.

**Type.** empirical  **Support.** evidenced

**Holds when.**

- requires a smooth and strongly convex loss function
- demonstrated across convex and nonconvex losses, time invariant and time varying topologies, and IID and non-IID datasets

**Source quote.**

> The DeceFL approach ensures that every client can reach the global minimum with zero performance gap and achieve the same convergence rate as centralized methods when the loss function is smooth and strongly convex.

**From.** Wulf A. Kaal, *AI Governance* (2024), Adapting Decentralization of AI Governance to AI Models, page 34

**Cite as.** Wulf A. Kaal, AI Governance (2024). SSRN: https://ssrn.com/abstract=4796714

**Verify.** sha256 of source PDF `59fa63bae179e8f9b6b8efbdf90cee28400276512a1b04f9f579a48641305c93` at https://raw.githubusercontent.com/wulfkaal/Academic-Papers/main/papers/pdf/Kaal%20-%202024%20-%20AI%20Governance.pdf

**Topics.** decentralization

**Keywords.** decefl, federated-learning, convergence, decentralization, performance

**Related claims.**

- restated_by: https://wulfkaal.github.io/claims/4941807-026

**Canonical form.** This markdown file is the canonical hashed representation of the claim. Its sha256 is the content hash used for attestation.