Gap-Free Compositions
- Gap-free compositions are sequences of positive integers summing to n where adjacent parts differ by at most one, ensuring a consecutive interval of part sizes.
- They are characterized by an explicit generating function, G(x) = (1+x)/(1-2x²-2x³), and a recurrence relation that reveals detailed combinatorial structure.
- This topic connects lattice path models and matrix reciprocity, providing foundational insights for bijective methods and probabilistic analysis in combinatorics.
A gap-free composition of a positive integer is a sequence of positive integers such that and for all . Equivalently, a composition is gap-free if the multiset of its part sizes forms a consecutive interval for some and every two adjacent parts differ by at most one. Gap-free compositions provide a natural intersection of the study of locally restricted compositions, lattice path models, and reciprocity in combinatorial matrix identities.
1. Formal Definitions and Initial Properties
Given a nonnegative integer , a composition of is said to have gap-bound if for all . The case is designated as gap-free: every pair of consecutive parts differs by at most one. The set of all gap-free compositions of is denoted by $\CA_{\leq 1}(n)$ or, equivalently, counted as $a_n=|\CA_{\leq 1}(n)|$ (Hopkins et al., 13 Dec 2025).
Small examples illustrate the structure. For :
- $4$
- $3+1$, $2+2$, $1+3$
- $3+1+1$, $2+2+1$, $1+2+2$, $2+1+1$, $1+2+1$, $1+1+3$
- Combinations such as $3+1+1+1$, $2+1+1+1$, and so on
The “gap-free” constraint aligns with a local restriction in the sense of Bender, Canfield, and Gao: if is gap-free, then for all . This property allows the combinatorial class to be embedded within the larger theory of locally restricted and asymptotically free compositions (Bender et al., 2012).
2. Generating Functions and Recurrence Relations
Gap-free compositions admit an explicit generating function and linear recurrence. The ordinary generating function is
which can also be written as
(Hopkins et al., 13 Dec 2025). The sequence satisfies the recurrence
The generating function can be motivated combinatorially by splitting compositions with respect to the parity of their length, analyzing adjacent pairs, and summing over patterns allowed by the gap-free condition. In particular, the structure can be built from atomic pairs and , extended by repeated concatenation and completed by possible odd-length extensions (Hopkins et al., 13 Dec 2025).
For bivariate enumeration counting both parts and total sum, the generating function is rational in both variables: where the coefficient of gives the number of gap-free compositions of with parts. The denominator reflects possible adjustments to adjacent part sizes (steps of ) (Beck et al., 2021).
3. Matrix Reciprocity and Combinatorial Involution
A foundational property of gap-free compositions is their appearance in a two-matrix inversion identity analogous to the Beck–Chern reciprocity for partitions and minimum-gaps (Beck et al., 2021). Define two infinite lower-triangular matrices:
- [q-binomial-like polynomial], interpreted as (signed) partition-type coefficients,
- , the number of gap-free compositions of into parts.
These matrices satisfy
where is the identity matrix. The proof employs an involution pairing between signed partitions and gap-free compositions, canceling contributions except for a unique diagonal fixed point . This involution corresponds to a stepwise decomposition: either move a part from the partition to the composition or vice versa within permitted constraints (Beck et al., 2021). This reciprocal structure elucidates the relationship between combinatorial and algebraic generating function techniques.
4. Asymptotics and Local Structure
The exponential growth rate of gap-free compositions is determined by the smallest positive real root of the equation
which evaluates numerically as , so for some (Hopkins et al., 13 Dec 2025). Standard singularity analysis applies via the location of this dominant singularity in the generating function.
Adopting the framework of locally restricted, asymptotically free compositions, random gap-free compositions display further structure:
- The mean largest part in a random gap-free composition of is , with precise corrections available via Mellin transform techniques.
- The expected number of distinct recurrent parts similarly grows as
- For fixed , the expected number of distinct recurrent parts of multiplicity remains bounded as
The fine enumerative structure is governed by asymptotic normality for small part counts and asymptotic geometric or Poisson behavior for large parts, allowing Poisson approximation methods for gap-freeness probabilities (Bender et al., 2012).
5. Probability of Gap-Freeness and Random Sampling
For a general recurrent, asymptotically free class , the probability that a random composition of is gap-free can be expressed as
where and depend on the combinatorial specification (Bender et al., 2012). The product form mirrors occupancy probabilities: a composition is gap-free if for each , there is at least one part of size . This is structurally analogous to the problem of a random sample from a geometric distribution omitting no elements of an initial segment.
This characterization enables the calculation of limiting distributions for maximal part size, number of distinct parts, and multiplicity counts, as well as explicit error bounds. For large , the bulk of the gap-freeness probability is concentrated near .
6. Connections to Carlitz–Arndt Compositions and Generalizations
Gap-free compositions are a central case in the broader “Carlitz–Arndt” class, in which the local constraint is imposed for some . The combinatorial and generating-function machinery specializes at to recover all classical gap-free results:
- Recurrence
- Ordinary generating function
- Bijections to subclasses (Hopkins et al., 13 Dec 2025)
Moreover, the Carlitz–Arndt framework facilitates extensions, such as colored parts, refined congruence constraints, and the connection to Rogers–Ramanujan-type partition identities. In particular, this setting clarifies the relationship between previously scattered enumeration results and opens structural interpretations for observed factorizations in generating functions.
7. Research Directions and Broader Context
Gap-free compositions continue to serve as a test case for new bijective methods, analytic combinatorics of locally restricted objects, and the development of matrix reciprocity. The link to asymptotically free local restrictions underpins a rich probabilistic and enumerative theory, informing studies of generalized pattern avoidance, universality of part-size statistics, and connections with lattice path enumerations.
These results have provided a laboratory for exploring the enumeration and random sampling of compositions under local conditions, setting the stage for research on more intricate local constraints and their impact on combinatorial structure and asymptotic behavior (Beck et al., 2021, Bender et al., 2012, Hopkins et al., 13 Dec 2025).