References & Further Reading
References
- M. A. Maddah-Ali and U. Niesen, Fundamental Limits of Caching, 2014
Contains the cut-set converse (Theorem 2) and the 12-gap order-optimality result. The reference for every converse argument in coded caching.
- Q. Yu, M. A. Maddah-Ali, and A. S. Avestimehr, The Exact Rate-Memory Tradeoff for Caching with Uncoded Prefetching, 2018
The headline converse result: MAN is exactly optimal under uncoded placement. The averaging argument is beautiful.
- K. Wan, D. Tuninetti, M. Ji, and G. Caire, Coded Caching with Correlated Files, 2020
CommIT contribution: file correlation exploited for rate reduction. Interference-alignment-based delivery. Chapter 12 of this book references this work as a template for extending CC to richer library structures.
- J. Gómez-Vilardebó, A Novel Centralized Coded Caching Scheme with Linear Coded Placement, 2018
First non-trivial example of coded placement strictly beating MAN for small $K$. Approximately 5-10% rate reduction. Foundational for the "coded placement beyond MAN" research program.
- C. Tian and J. Chen, Caching and Delivery via Interference Elimination, 2018
Proves that under symmetric coded placements, MAN remains optimal. Complements YMA '18 by extending the optimality to a broader class.
- Z. Chen, P. Fan, and K. B. Letaief, Fundamental Limits of Caching: Improved Bounds For Small Buffer Users, 2016
Improves the MN '14 converse for the small-cache regime ($M \leq N/K$). Used as a building block in subsequent works.
- C.-Y. Wang, S. S. Bidokhti, and M. Wigger, Improved Converses and Gap Results for Coded Caching, 2018
Refined converse bounds using genie-aided arguments and new entropy inequalities. Narrows the MAN-vs-converse gap below 12.
- H. Ghasemi and A. Ramamoorthy, Improved Lower Bounds for Coded Caching, 2017
LP-based converse bounds. Tighter than cut-set for specific $(K, N, M)$, but computational.
- S. Jin, Y. Cui, H. Liu, and G. Caire, Structural Properties of Uncoded Placement Optimization for Coded Delivery, 2017
Caire group's analysis of optimal uncoded placement under non-uniform demand. Connects Chapter 3's worst-case theory with Chapter 13's non-uniform analysis.
- M. Ji, A. M. Tulino, J. Llorca, and G. Caire, Order-Optimal Rate of Caching and Coded Multicasting with Random Demands, 2017
Order-optimal bounds for random (Zipf) demands. Bridges worst-case and average-case theories.
- T. M. Cover and J. A. Thomas, Elements of Information Theory, Wiley-Interscience, 2nd ed., 2006
For Fano's inequality, the chain rule, and the information-theoretic tools used in the converse proofs of this chapter.
- A. El Gamal and Y.-H. Kim, Network Information Theory, Cambridge University Press, 2011
For cut-set bounds in the broadcast channel and network coding contexts — the parent theory for coded-caching converses.
- K. Wan and G. Caire, On Coded Caching with Private Demands, 2021
CommIT demand-privacy result (Chapter 12). Shows that demand privacy can be achieved at the same rate as MAN — remarkable that privacy is "free." Referenced here because of the connection to the converse question: does privacy change the fundamental rate?
Further Reading
Resources for students who want to master the coded-caching converse literature before Chapter 4.
The YMA '18 paper
Yu, Maddah-Ali, Avestimehr, 'Exact Rate-Memory Tradeoff...' IEEE Trans. IT, 2018
The definitive converse under uncoded placement. The averaging argument over demand types is the key technique.
Converse bounds survey
Wang, Lin, Sun, Bidokhti, Wigger: 'Tight bounds for coded caching' IEEE Trans. IT, 2018
Refined converses beyond MN '14. Tightens the order-optimality from 12 to smaller constants.
Open problems perspective
Maddah-Ali & Niesen, 'Coding for Caching' (tutorial), IEEE ComSoc 2016
Accessible overview of the coded-caching open problems and research directions. Good companion before tackling the research literature.
Correlated files theory
Wan, Tuninetti, Ji, Caire: 'Coded Caching with Correlated Files' IEEE Trans. IT, 2020
The interference-alignment-based delivery scheme and its analysis. Template for extending MAN to structured libraries.