How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Relaxed inequalities for squared Euclidean norms
Statement
For vectors and , For vectors ,
Facts & Assumptions
Given: The standard Euclidean inner product and norm on , with .
The real inner product is symmetric and bilinear, and squared norms are nonnegative.
Proof
Expanding the first inequality's right side minus its left side gives [given] Replacing by gives the same bound with ; together these two bounds yield both displayed two-vector inequalities.
Expanding each squared difference and collecting diagonal and cross terms [given] gives This proves the finite-sum inequality, including , when the right difference is zero.
Used by
- Client drift under FedAvg gradient dissimilarity Lemma
- Convex SCAFFOLD client-drift bound Lemma
- Convex SCAFFOLD control-lag recursion Lemma
- Convex SCAFFOLD server-update second moment Lemma
- Deviation of local-SGD workers from their virtual mean Lemma
- Nonconvex SCAFFOLD control-anchor tracking recursion Lemma
- Nonconvex SCAFFOLD progress in one communication round Lemma
- Nonconvex SCAFFOLD query-point drift bound Lemma
- Nonconvex SCAFFOLD server-update moments Lemma
- One-round progress of strongly convex FedAvg Lemma
- One-step progress of the virtual local-SGD iterate Lemma
- Strong-convexity inequality for a perturbed gradient point Lemma
Dependency tree · 0 levels
Nothing. This result depends on no other item in the library.
Sources
- Sai Praneeth Karimireddy et al., SCAFFOLD, Appendix B, Lemma 3 (standard reference, not scraped)