Online Learning: A Modern Introduction Using Convex Optimization
Francesco Orabona
arXiv:1912.13213v10, June 21, 2026; final preprint
Bibliographic metadata checked 2026-09-09. Page and theorem mappings are separately audited.Planned reading map
Online Learning Book
Convex optimization and regret minimization. Existing EXP3/FTRL routes are shared references; adjacent online-learning frontier questions are indexed separately from core Bandit/RL open problems.
Existing shared reading
These links reuse established pages with their original sources and exact Lean boundaries. They do not certify a chapter of the new book.
Planned source mapping
Next: freeze source versions, chapter contracts, assumptions and theorem locators; retrieve existing declarations and prove only the missing interfaces. Chapter numbers, page coverage and completion totals will appear after that audit.