Statistics on Graphs, Exponential Formula and Combinatorial Physics

The concern of this paper is a famous combinatorial formula known under the name “exponential formula”. It occurs quite naturally in many contexts (physics, mathematics, computer science). Roughly speaking, it expresses that the exponential generating function of a whole structure is equal to the exponential of those of connected substructures. Keeping this descriptive statement as a guideline, we develop a general framework to handle many different situations in which the exponential formula can be applied.

💡 Research Summary

The paper revisits the celebrated exponential formula—a cornerstone of combinatorial enumeration that links the exponential generating function (EGF) of a class of combinatorial structures to the exponential of the EGF of its connected components. While the classical statement is well‑known for labelled objects such as graphs, set partitions, or permutations, the authors aim to place this identity in a far more abstract and versatile setting, thereby exposing its underlying algebraic and categorical mechanisms and demonstrating its applicability across mathematics, computer science, and theoretical physics.

The exposition begins with a concise historical review, recalling how the formula first appeared in the work of Touchard, Riddell, and later in the context of species by Joyal. The authors stress that the “connected” notion is not limited to graph connectivity; it is any decomposition of an object into irreducible, non‑splittable pieces under a suitable binary operation (typically disjoint union). This viewpoint motivates the introduction of a general category 𝒞 whose objects are the combinatorial structures of interest and whose morphisms preserve the underlying labeling or symmetry. Within 𝒞, a distinguished full subcategory 𝒞_conn collects the connected objects.

A central hypothesis—called the “unique multiset decomposition property”—requires that every object A∈𝒞 can be written uniquely (up to isomorphism) as a multiset of connected objects. Formally, A≅⊎_{i∈I} C_i with C_i∈𝒞_conn, and the indexing multiset I is uniquely determined. The authors then attach a weight function w:C_conn→ℚ