Table of Contents
Fetching ...

Deterministic approximation for the volume of the truncated fractional matching polytope

Heng Guo, Vishvajeet N

TL;DR

This work develops a deterministic FPTAS for the volume of the truncated fractional matching polytope on graphs of maximum degree $\Delta$ (with $\delta\le C/\Delta$) and extends the result to hypergraphs with maximum edge size $k$ (with $\delta\le C\Delta^{-(2k-3)/(k-1)}k^{-1}$). The core technique is an algorithmic cluster expansion: the volume is expressed as a polymer partition function, the Kotecký–Preiss criterion is verified to ensure convergence, and weights are computed via polynomial-time polytope-volume calculations. The method generalises prior work on the truncated independence polytope and BR24, and relies on a careful polymer formulation and a hypergraph-specific MCS-based model to achieve the necessary decay. The results provide explicit truncation-parameter bounds under which a deterministic approximation is feasible, while leaving open the challenge of reducing these bounds further or removing truncation altogether.

Abstract

We give a deterministic polynomial-time approximation scheme (FPTAS) for the volume of the truncated fractional matching polytope for graphs of maximum degree $Δ$, where the truncation is by restricting each variable to the interval $[0,\frac{1+δ}Δ]$, and $δ\le \frac{C}Δ$ for some constant $C>0$. We also generalise our result to the fractional matching polytope for hypergraphs of maximum degree $Δ$ and maximum hyperedge size $k$, truncated by $[0,\frac{1+δ}Δ]$ as well, where $δ\le CΔ^{-\frac{2k-3}{k-1}}k^{-1}$ for some constant $C>0$. The latter result generalises both the first result for graphs (when $k=2$), and a result by Bencs and Regts (2024) for the truncated independence polytope (when $Δ=2$). Our approach is based on the cluster expansion technique.

Deterministic approximation for the volume of the truncated fractional matching polytope

TL;DR

This work develops a deterministic FPTAS for the volume of the truncated fractional matching polytope on graphs of maximum degree (with ) and extends the result to hypergraphs with maximum edge size (with ). The core technique is an algorithmic cluster expansion: the volume is expressed as a polymer partition function, the Kotecký–Preiss criterion is verified to ensure convergence, and weights are computed via polynomial-time polytope-volume calculations. The method generalises prior work on the truncated independence polytope and BR24, and relies on a careful polymer formulation and a hypergraph-specific MCS-based model to achieve the necessary decay. The results provide explicit truncation-parameter bounds under which a deterministic approximation is feasible, while leaving open the challenge of reducing these bounds further or removing truncation altogether.

Abstract

We give a deterministic polynomial-time approximation scheme (FPTAS) for the volume of the truncated fractional matching polytope for graphs of maximum degree , where the truncation is by restricting each variable to the interval , and for some constant . We also generalise our result to the fractional matching polytope for hypergraphs of maximum degree and maximum hyperedge size , truncated by as well, where for some constant . The latter result generalises both the first result for graphs (when ), and a result by Bencs and Regts (2024) for the truncated independence polytope (when ). Our approach is based on the cluster expansion technique.
Paper Structure (8 sections, 11 theorems, 38 equations)

This paper contains 8 sections, 11 theorems, 38 equations.

Key Result

Theorem 3

For graphs of maximum degree $\Delta\ge 2$ and $\delta\le\frac{C}{\Delta}$ for some constant $C>0$, there is a fully polynomial-time approximation scheme (FPTAS) for $\operatorname{Vol}(P_{G,\delta})$.

Theorems & Definitions (23)

  • Definition 1: Fractional matching polytope
  • Definition 2: Truncated fractional matching polytope
  • Theorem 3
  • Theorem 4
  • Definition 5: Polymer partition function
  • Definition 6: Cluster Expansion
  • Proposition 7: Kotecký-Preiss Criterion KP86
  • Proposition 8: JKP20
  • Lemma 9: BCKL13
  • Lemma 10
  • ...and 13 more