Channel capacities and open gaps.

A community-maintained registry of channel capacities, open gaps, and Lean statements.

32Problems 8Open
6Lean statements
0External proofs

Featured open problems

Each input bit is independently deleted without an erasure marker; the exact capacity is unknown for every nontrivial deletion probability.

Point-to-point Binary Finite alphabet Memory Deletion Capacity Bounds only
Open \(0.1221(1-d)<C_{\mathrm{del}}(d)\le0.3578(1-d)\)

A causal relay assists a source, but decode-forward and the cut-set bound do not coincide in general.

Relay Finite alphabet Discrete memoryless Capacity Bounds only
Open \(R_{\mathrm{DF}}\le C\le R_{\mathrm{cut}}\)

The capacity region for two arbitrary broadcast receivers remains unknown outside important ordered subclasses.

Broadcast Finite alphabet Discrete memoryless Capacity region Bounds only
Open \(\mathcal R_{\mathrm{Marton}}\subseteq\mathcal C_{\mathrm{BC}}\subseteq\mathcal R_{\mathrm{UV}}\)

Two transmitter-receiver pairs interfere, and the exact capacity region is unknown in general.

Interference Finite alphabet Discrete memoryless Capacity region Bounds only
Open \(\mathcal R_{\mathrm{HK}}\subseteq\mathcal C_{\mathrm{IC}}\subseteq\mathcal R_{\mathrm{outer}}\)

A multiple-unicast instance where the exact nonlinear symmetric capacity is separated from the known linear answer.

Index coding Finite alphabet Multiple unicast Non-Shannon inequalities Nonlinear coding Symmetric capacity Bounds only Linear-only result
Open \(\frac5{13}\le C_{\mathrm{sym}}\le\frac{11}{28}\)

For a finite confusability graph, the regularized independence number defines capacity but is difficult to compute or characterize.

Zero error Finite alphabet Discrete memoryless Zero-error capacity Regularized characterization Bounds only
Open \(\Theta(G)=\sup_{n\ge1}\alpha(G^{\boxtimes n})^{1/n}\)

A registry, not a proof monorepo

Capacity Atlas keeps canonical channel specifications, controlled tags, and versioned Lean statements in one reviewable repository. Substantial proofs live in dedicated repositories and link back by immutable commit.

The statement-first organization, external-proof links, and simple browse interface are inspired by Google DeepMind's Formal Conjectures. See the acknowledgements for the exact design influences and citation.