Combinatorial properties of holographic entropy inequalities

Technical Summary: Combinatorial Properties of Holographic Entropy Inequalities

Problem Statement
Holographic entropy inequalities (HEIs) are linear constraints satisfied by Ryu-Takayanagi (RT) holographic entanglement entropies, which correspond to minimum cut functions on weighted graphs. While the structure of these inequalities is known for small numbers of parties ($N \le 5$) and partially for $N=6$, a general combinatorial characterization for arbitrary $N$ remains elusive. A specific class of HEIs, known as superbalanced inequalities, was the subject of a previous study [1], which proposed four conjectures regarding their properties:

1. Conjecture 1: If an inequality is an HEI, all its null reductions obey a specific majorization property.
2. Conjecture 2: If all null reductions of an inequality obey the majorization property, the inequality is an HEI.
3. Conjecture 3: If an inequality is an HEI, all its null reductions are HEIs.
4. Conjecture 4: If all null reductions of an inequality are HEIs, the inequality is an HEI.

The paper aims to resolve these conjectures and establish a new combinatorial framework to characterize HEIs, particularly focusing on centered inequalities (those where a specific party appears in every term).

Methodology
The authors employ a matrix-based formalism where an inequality is represented by a pair of $(0,1)$-matrices, $L$ and $R$, encoding the occurrence of parties in the left-hand side (LHS) and right-hand side (RHS) terms, respectively. The study introduces and analyzes several new combinatorial properties defined on these matrix pairs:

• Dominance Properties:
  • Dominance: For every subset of LHS terms, there exists an equal-sized subset of RHS terms that "dominates" it (contains every party at least as many times).
  • Inclusion Dominance: A stronger condition requiring that the dominance map preserves inclusion relations between subsets of terms.
  • Region Dominance: A condition comparing the number of terms containing a specific collection of parties (a "region") on the LHS versus the RHS.
• Majorization Properties:
  • Positive-Combinations Majorization (PCM): $L \preceq_{PC} R$ if $vL \preceq vR$ for all non-negative real vectors $v$.
  • Binary-Combinations Majorization (BCM): $L \preceq_{BC} R$ if $vL \preceq vR$ for all binary vectors $v \in \{0,1\}^m$.
• Contraction Maps: The existence of a map $f: \{0,1\}^{m_L} \to \{0,1\}^{m_R}$ that preserves the occurrence vectors of parties and non-expands the Hamming distance. The paper assumes the equivalence between the existence of such a map and an inequality being an HEI.

The methodology involves proving logical implications between these properties, specifically for centered inequalities (derived from null reductions of superbalanced inequalities), and constructing explicit counterexamples to disprove converse implications.

Key Contributions and Results

1. Resolution of Conjectures from [1]:

   • Conjecture 1 (Proven): The authors prove that if an inequality is an HEI, all its null reductions satisfy the majorization property (specifically PCM). This is established via Theorem 1 (null reduction of superbalanced HEIs yields HEIs) and Theorem 3 (dominance implies PCM).
   • Conjecture 2 (Disproven): The paper provides a direct counterexample (an $N=6$ superbalanced non-HEI) where all null reductions obey PCM, yet the original inequality is not an HEI.
   • Conjecture 3 (Proven): It is proven that the null reduction of any superbalanced HEI is itself an HEI (Theorem 1).
   • Conjecture 4 (Disproven): A counterexample is provided where all null reductions of a superbalanced inequality are HEIs, but the original inequality is not. The authors propose a weaker version, Conjecture 4', involving null reductions with varied purifiers, which remains viable but unproven.

2. Equivalence of Contraction and Inclusion Dominance:

   • Theorem 2: For centered inequalities, the existence of a contraction map is equivalent to the existence of an inclusion dominance map. This establishes that inclusion dominance is a necessary and sufficient combinatorial condition for a centered inequality to be an HEI.

3. Hierarchy of Combinatorial Properties:
   The paper establishes a strict hierarchy of implications for centered inequalities:
   $$\text{HEI} \iff \text{Inclusion Dominance} \implies \text{Dominance} \implies \text{PCM} \implies \text{BCM} \implies \text{Region Dominance}$$
   The paper provides counterexamples for every reverse implication, demonstrating that these properties are distinct and that the hierarchy is strict.

4. Generalization to Balanced HEIs:
   Theorem 5 proves that any balanced HEI (not necessarily centered) satisfies region dominance. This is proven directly using the min-cut function on star graphs, independent of contraction maps.

Significance and Claims
The paper claims to resolve the open questions posed in [1] regarding the relationship between null reductions, majorization, and the holographic entropy cone (HEC).

• Evidence for Time-Dependent States: By proving Conjecture 1, the authors provide strong evidence that time-dependent holographic states (governed by the covariant HRT formula) obey the same entropy inequalities as static states. The null reduction test, which is sensitive to time-dependent configurations, is shown to be a necessary condition for any HEI.
• Combinatorial Characterization: The work establishes that the complex geometric problem of determining HEIs can be translated into a purely combinatorial problem involving matrix dominance and majorization. Specifically, the equivalence between contraction maps and inclusion dominance offers a concrete, checkable criterion for identifying HEIs among centered inequalities.
• Limitations of Majorization: The disproof of Conjecture 2 clarifies that while majorization properties are necessary, they are not sufficient to characterize the full set of HEIs. The "bootstrap" approach of inferring general HEIs solely from their light-cone (null-reduced) configurations is insufficient without additional constraints.

The authors conclude that these combinatorial properties offer a new structural perspective on the holographic entropy cone, potentially aiding in the systematic generation of new inequalities and the understanding of the multipartite entanglement structure in quantum gravity.