Enumeration of Odd Dimensional Partitions modulo 4

Basic Information

• Paper ID: 2207.07513
• Title: Enumeration of Odd Dimensional Partitions modulo 4
• Author: Aditya Khanna
• Classification: math.CO (Combinatorics)
• Submission Date: First submitted July 2022, latest version (v3) November 15, 2025
• Paper Link: https://arxiv.org/abs/2207.07513

Abstract

This paper investigates the enumeration of integer partitions by dimension modulo 4. The dimension $f^λ$ of a partition λ is defined as the number of standard Young tableaux of its corresponding shape. The author defines $a_i(n)$ as the number of partitions of n with dimension congruent to i modulo 4, and refines results of Macdonald and McKay on odd-dimensional partitions. Specifically, explicit formulas are computed for $a_1(n)$ and $a_3(n)$ when n's binary expansion contains no consecutive 1's or has binary digit sum equal to 2.

Research Background and Motivation

Research Problem

The core problem addressed is: How can one precisely count the number of integer partitions with dimension congruent to 1 or 3 modulo 4? Defining $\delta(n) = a_1(n) - a_3(n)$, the research goal is to find explicit formulas for δ(n) for specific values of n.

Problem Significance

1. Connection to Symmetric Group Representation Theory: Partitions of integer n index irreducible representations of the symmetric group $S_n$, with partition dimension corresponding to representation dimension. Modulo 4 properties are directly related to the classification of spinor representations of symmetric groups.
2. Refinement of Existing Results: Macdonald used p-core tower theory to compute $m_2(n)$ (the number of odd-dimensional partitions). For $n = 2^{k_1} + \cdots + 2^{k_\ell}$, he obtained $m_2(n) = 2^{k_1+\cdots+k_\ell}$. This paper refines this result to the modulo 4 level.
3. Representation Theory Applications: Previous research has shown that irreducible representations corresponding to odd partitions have special properties under restriction, with zero density of odd character values. The refined modulo 4 structure helps understand deeper properties of these representations.

Limitations of Existing Methods

• Macdonald's results only give the total number of odd-dimensional partitions, without distinguishing between residues 1 and 3 modulo 4
• Work by Amrutha P and T. Geetha computes $m_4(2^\ell)$ and $m_8(2^\ell)$, but does not provide complete characterization for general n
• Lack of systematic methods for handling the recursive structure of modulo 4 counting

Research Motivation

The author introduces the concept of "parent partitions" and the Od function (extracting the odd part modulo 4), establishing recursive relationships for the modulo 4 properties of dimensions, thereby enabling treatment of n values with specific binary structures.

Core Contributions

1. Complete Solution for Sparse Numbers (Theorem 2): For sparse numbers (binary expansion with no consecutive 1's), explicit formulas for $\delta(n)$ are provided:
   • $\delta(2) = 2$
   • $\delta(n) = 0$ (n > 2 even)
   • $\delta(n) = 4^{\nu(n)-1}$ (n odd), where ν(n) is the number of 1's in the binary expansion
2. Recursive Formula (Theorem 1): For $n = 2^R + m$ with $2^{R-1} > m$, a recursive relationship is established: $\delta(n) = \begin{cases} 0, & \text{if } n \text{ is even} \\ 4\delta(m), & \text{if } n \text{ is odd} \end{cases}$
3. Case of Binary Sum Equal to 2 (Theorem 3): For $n = 2^R + 2^{R-1}$, it is proven that: $\delta(n) = \begin{cases} 2, & R=1 \\ 8, & R=2 \\ 0, & \text{otherwise} \end{cases}$
4. Technical Innovations:
   • Introduction of the Od function to systematically handle the odd part of dimension modulo 4
   • Establishment of the "workhorse formula" (Theorem 36) linking dimensions of partitions and their $2^R$-parent partitions
   • Unified treatment of hook removal operations through β-set techniques

Detailed Methodology

Task Definition

Input: Positive integer n Output: $\delta(n) = a_1(n) - a_3(n)$, the difference between the number of partitions with dimension ≡ 1 (mod 4) and those with dimension ≡ 3 (mod 4) Constraints: This paper focuses on cases where n satisfies specific binary structures

Core Conceptual Framework

1. β-Sets and Hook Operations

For a partition λ, define the first column hook length set $H(\lambda) = \{h_{i,1} | 1 \leq i \leq \ell(\lambda)\}$, where $h_{i,1} = \lambda_i + k - i$.

The β-set is defined as $H(\lambda)^{+r} = \{x+r | x \in H(\lambda)\} \cup \{0,\ldots,r-1\}$.

Proposition 15: λ contains a t-hook if and only if there exists $h \in H(\lambda)$ such that $h \geq t$ and $h-t \notin H(\lambda)$. The resulting partition μ after removal satisfies $H(\mu) \sim_\beta H(\lambda)[h \to h-t]$.

2. Characterization of Odd Partitions

Proposition 21 (Macdonald): Let $n = 2^R + m$ with $m < 2^R$. Then λ ⊢ n is an odd partition if and only if λ contains exactly one $2^R$-hook and $\text{core}_{2^R}(\lambda)$ is also an odd partition.

Definition 22: If $\text{core}_{2^R}(\lambda) = \mu$, then λ is called a $2^R$-parent partition of μ.

3. Classification of Parent Partitions

Proposition 23: The $2^R$-parent partitions of μ fall into exactly two classes:

• Type I: $H(\lambda) = H(\mu)[x \to x+2^R]$, for some $x \in H(\mu)$
• Type II: $H(\lambda) = H(\mu)^{+r}[0 \to 2^R]$, $1 \leq r \leq 2^R$ and $2^R \notin H(\mu)^{+r}$

There are $\ell(\mu)$ Type I parent partitions, $2^R - \ell(\mu)$ Type II parent partitions, for a total of $2^R$.

4. Od Function

Definition 27: For $n \in \mathbb{N}$, let $v_2(n)$ be the largest power of 2 dividing n. Define $\text{Od}(n) = \begin{cases} 1, & \text{if } n/2^{v_2(n)} \equiv 1 \pmod{4} \\ -1, & \text{if } n/2^{v_2(n)} \equiv 3 \pmod{4} \end{cases}$

Lemma 28: The Od function is multiplicative, i.e., $\text{Od}(mn) = \text{Od}(m)\text{Od}(n)$.

Lemma 33: For factorials, there is an explicit formula $\text{Od}(n!) = (-1)^{D(n)+\nu(\lfloor n/4 \rfloor)}$ where $D(n)$ is the number of consecutive pairs of 1's in the binary expansion of n.

Core Technique: Workhorse Formula

Definition of η Statistic

Definition 34: Let λ be a $2^R$-parent partition of μ, and $h_\mu^\lambda$ be the affected hook length. Define $(-1)^{\eta_\mu^\lambda} = \prod_{\substack{x \in H(\lambda) \\ x \neq h_\mu^\lambda}} \frac{\text{Od}(|h_\mu^\lambda - x|)}{\text{Od}(|h_\mu^\lambda - 2^R - x|)}$

Proposition 39: An explicit computation formula for η is $\eta_\mu^\lambda = N_\lambda(h_\mu^\lambda) - I_{H(\lambda)}(h_\mu^\lambda - 2^{R-1}) + I_{H(\lambda)}(h_\mu^\lambda + 2^{R-1}) + I_{H(\lambda)}(h_\mu^\lambda - 3 \cdot 2^{R-1})$ where $N_\lambda(h) = |\{y \in H(\lambda) | h-2^R < y < h\}|$.

Workhorse Formula

Proposition 36: Let $n = 2^R + m > 3$ with $m < 2^R$, and λ be a $2^R$-parent partition of μ. Then $\text{Od}(f^\lambda) = (-1)^{s_2(n)+s_2(h_\mu^\lambda)+\eta_\mu^\lambda} \text{Od}(f^\mu)$ where $s_2(n)$ is the sum of the first two binary digits of n.

Proof Strategy:

1. Use the Frobenius formula to compute dimension from β-sets: $f^\lambda = \frac{n! \prod_{1 \leq i < j \leq k}(h_i - h_j)}{\prod_{i=1}^k h_i!}$
2. Apply the Od function and Lemma 33 to handle factorials
3. Analyze the difference between $H(\lambda)$ and $H(\mu)^{+r}$, utilizing the sign changes produced by the jump of $h_\mu^\lambda$

Simplification for Sparse Numbers

Corollary 47: When $m < 2^{R-1}$ (corresponding to sparse numbers), the workhorse formula simplifies to $\text{Od}(f^\lambda) = (-1)^{\eta_\mu^\lambda} \text{Od}(f^\mu)$ since $s_2(n) = s_2(h_\mu^\lambda) = 1$.

Corollary 48: In this case, the η formula simplifies to $\eta_\mu^\lambda = N_\lambda(h_\mu^\lambda) - I_{H(\lambda)}(h_\mu^\lambda - 2^{R-1})$

Parent Partition Counting

Introduce the signed sum: $S_\Lambda(\mu) = \frac{1}{\text{Od}(f^\mu)} \sum_{\lambda \in \Lambda} \text{Od}(f^\lambda)$

Proposition 51 (Type I): For an odd partition μ with $m < 2^{R-1}$, $S_{P_1(\mu)}(\mu) = \begin{cases} 0, & \text{if } \ell(\mu) \text{ is even} \\ 1, & \text{if } \ell(\mu) \text{ is odd} \end{cases}$

Proposition 53 (Type II): $S_{P_2(\mu)}(\mu) = \begin{cases} 2 - 2(-1)^m, & \text{if } \ell(\mu) \text{ is even} \\ 1 - 2(-1)^m, & \text{if } \ell(\mu) \text{ is odd} \end{cases}$

The proof of Type II requires introducing the parity gap statistic $G(X) = \sum_{x \in X} (-1)^x$ and analyzing two subsets of $P_2(\mu)$: $P_2^\uparrow(\mu)$ (where $r \leq 2^{R-1}$) and $P_2^\downarrow(\mu)$ (where $r > 2^{R-1}$).

Lemma 59: For an odd partition λ, $G(H(\lambda)) = \begin{cases} 1-(-1)^n, & \text{if } \ell(\lambda) \text{ is even} \\ (-1)^n, & \text{if } \ell(\lambda) \text{ is odd} \end{cases}$

Establishment of Recursive Formula

Combining results for Type I and Type II: $S_{P(\mu)}(\mu) = S_{P_1(\mu)}(\mu) + S_{P_2(\mu)}(\mu) = 2 - 2(-1)^m$

Therefore: $\delta(n) = \sum_{\mu \vdash m, \mu \text{ odd}} S_{P(\mu)}(\mu) \text{Od}(f^\mu) = (2-2(-1)^m) \delta(m)$

This yields the recursive relationship of Theorem 1.

Experimental Setup

Data Generation

This is a pure theoretical mathematics paper; "experiments" refer to numerical verification:

• Computed all $\delta(n)$ values for $n=0$ to $n=127$ (see appendix table)
• Used the hook-length formula to compute partition dimensions
• Employed β-set algorithms to enumerate partitions and their parent partitions

Verification Methods

• For small n (n ≤ 15), directly enumerated all partitions and computed dimensions
• For larger n, used recursive formulas and compared with direct computation
• Verified theoretical predictions for special values (e.g., the $4^{\nu(n)-1}$ pattern for sparse numbers)

Experimental Results

Main Result Verification

Verification for Sparse Numbers

From appendix data, Theorem 2 can be verified:

• $\delta(1) = 1 = 4^{1-1}$ ✓
• $\delta(2) = 2$ ✓
• $\delta(5) = 4 = 4^{2-1}$ (5=101₂, ν(5)=2) ✓
• $\delta(9) = 4 = 4^{2-1}$ (9=1001₂, ν(9)=2) ✓
• $\delta(17) = 4 = 4^{2-1}$ (17=10001₂, ν(17)=2) ✓
• $\delta(21) = 16 = 4^{3-1}$ (21=10101₂, ν(21)=3) ✓
• $\delta(41) = 16 = 4^{3-1}$ (41=101001₂, ν(41)=3) ✓

Even sparse numbers:

• $\delta(4) = 0$, $\delta(8) = 0$, $\delta(16) = 0$, $\delta(32) = 0$ ✓

Verification for Binary Sum Equal to 2

Theorem 3 predicts for $n = 2^R + 2^{R-1}$:

• $n=3$ (R=1): $\delta(3) = 2$ ✓
• $n=6$ (R=2): $\delta(6) = 8$ ✓
• $n=12$ (R=3): $\delta(12) = 0$ ✓
• $n=24$ (R=4): $\delta(24) = 0$ ✓
• $n=48$ (R=5): $\delta(48) = 0$ ✓
• $n=96$ (R=6): $\delta(96) = 0$ ✓

Complexity of Non-Sparse Numbers

Appendix data reveals the complexity of the case $2^{R-1} < m < 2^R$:

Example: $n=118 = 64 + 54 = 2^6 + 54$ ($54 > 2^5$)

• $\delta(118) = -384$, neither a power of 2 nor positive

Dramatic Fluctuations in Consecutive Intervals (n=122 to 127):

n=122: δ=-256
n=123: δ=-256
n=124: δ=768
n=125: δ=640
n=126: δ=168
n=127: δ=256

This indicates that in the unresolved case, δ(n) behaves highly irregularly, depending on the fine structure of partitions.

Key Findings

1. Breakdown of Power-of-2 Pattern: For non-sparse numbers, δ(n) is no longer a power of 2, and can even be negative
2. Dependence on Core Structure: Calculations show that $S_{P(\mu)}(\mu)$ no longer depends only on |μ|, but on the specific structure of μ
3. Increased Recursion Depth: Unresolved cases may require consideration of $2^{R-1}$-cores or even deeper recursive levels

Related Work

Enumeration of Odd Partitions

1. McKay's Work: First systematic study of odd-dimensional partitions
2. Macdonald 1: Used p-core tower theory to obtain the elegant formula $m_2(n) = 2^{k_1+\cdots+k_\ell}$
3. Ayyer, Prasad, Spallone 5: Studied structural properties of odd partitions in Young's lattice

Research on Higher Powers Modulo

1. Amrutha P and T. Geetha 3:
   • Provided general recursive results for $m_{2^k}(n)$
   • Computed $m_4(2^\ell)$ and $m_8(2^\ell)$
   • Gave characterization of partitions of $2^\ell$ with dimension ≡ 2 (mod 4)
2. Comparison with This Paper: More refined distinction between the two odd residue classes modulo 4, but only resolves n with specific binary structures

Representation Theory Background

1. Ganguly and Spallone 2: Spinor representation classification provides representation-theoretic motivation for studying modulo 4
2. Giannelli et al. 4: Restriction properties of odd-degree characters
3. Peluse 6, Ganguly et al. 7: Density results for character values modulo primes

Combinatorial Tools

1. Olsson 8: Systematic exposition of p-core and quotient theory
2. Macdonald 9: Symmetric functions and Hall polynomials, β-set techniques
3. Eğecioğlu and Remmel 12: Combinatorial interpretation of inverse Kostka matrices, special rim hook concept

Innovations in This Paper

• First systematic computation of the difference $a_1(n) - a_3(n)$
• Introduction of the Od function as a technical tool
• Establishment of the workhorse formula linking partitions at different levels
• Complete resolution of the sparse number case

Conclusions and Discussion

Main Conclusions

1. Complete Resolution for Sparse Numbers: For n whose binary expansion contains no consecutive 1's, δ(n) has an explicit formula; for odd n, it equals $4^{\nu(n)-1}$
2. Establishment of Recursive Framework: Through parent partition theory and the workhorse formula, a recursive structure for modulo 4 counting is established
3. Treatment of Special Cases: For $n = 2^R + 2^{R-1}$, δ(n) is nonzero only when R=1,2

Limitations

The author candidly discusses the limitations of the method in Section 6:

1. Unresolved Region: For $2^{R-1} < m < 2^R$, $S_{P(\mu)}(\mu)$ depends on the specific structure of μ, preventing unified treatment
2. Technical Obstacles:
   • The injective map $h \mapsto r_h$ in Lemma 58 no longer holds
   • Additional conditions like checking $3 \cdot 2^{R-1} \in H(\mu)$ are needed
   • Computation of Type I parent partitions also becomes dependent on elements of $H(\mu)$
3. Increased Recursion Depth: May require consideration of $2^{R-1}$-cores or deeper levels, significantly increasing complexity
4. Numerical Evidence: δ(118) = -384 shows that values are no longer powers of 2, and can be negative and large

Future Directions

The paper poses three open problems:

1. Main Problem: Compute δ(n) for $n = 2^R + m$ with $2^{R-1} < m < 2^R$
2. Bounding Problem: Can one give reasonable bounds for $S_{P(\mu)}(\mu)$?
3. Characterization Problem: Does there exist a method to directly characterize modulo 4 odd partitions using β-sets?

The author suggests that progress may require:

• Development of new combinatorial tools for handling multi-level core structures
• Search for new invariants of β-sets capturing modulo 4 properties
• Computer-assisted exploration of patterns

In-Depth Evaluation

Strengths

1. Innovation of Methods

• Introduction of Od Function: Elegantly formalizes the "odd part modulo 4" operation; its multiplicative property (Lemma 28) makes it a powerful tool
• Workhorse Formula: Proposition 36 establishes precise connections between dimensions of partitions at different levels, serving as the technical core of the paper
• Systematic Classification of Parent Partitions: The Type I/II classification (Proposition 23) provides clear combinatorial structure

2. Rigor of Proofs

• Every technical lemma has complete proofs with clear logical chains
• Lemma 33 (formula for Od(n!)) demonstrates sophisticated binary analysis techniques
• Proposition 39 (explicit formula for η) carefully considers three cases

3. Elegance of Results

• Theorem 2 gives the formula $\delta(n) = 4^{\nu(n)-1}$ (for odd sparse numbers) in simple form
• Forms a beautiful refinement of Macdonald's $m_2(n) = 2^{k_1+\cdots+k_\ell}$
• Recursive formula (Theorem 1) has clear structure and is easy to apply

4. Writing Quality

• Section 2 systematically reviews background (partitions, hooks, cores, etc.) with strong self-containment
• Numerous examples (e.g., Examples 17, 25, 41) aid understanding of abstract concepts
• Appendix provides complete data for n=0 to 127, facilitating verification and exploration

Weaknesses

1. Limited Scope of Applicability

• Only resolves sparse numbers and cases with $\nu(n)=2$; general n remains unresolved
• From the data, unresolved cases constitute the majority (e.g., n=29,30,31, etc.)
• Section 6 acknowledges that the case $2^{R-1} < m < 2^R$ is "so difficult that it evades our methods"

2. High Technical Complexity

• Proofs of Lemmas 58 and 61 involve multi-level summations and indicator functions, requiring sophisticated techniques but difficult to generalize
• Introduction of parity gap statistic G(X) appears somewhat ad hoc
• Computation of η (Proposition 39) requires multiple case distinctions, lacking a unified framework

3. Insufficient Theoretical Explanation

• Why are sparse number cases particularly simple? The underlying deep reasons are not fully elucidated
• What are the combinatorial or representation-theoretic meanings of the sign and magnitude of δ(n)?
• The specific connection to spinor representation classification is only mentioned in the introduction, not deeply developed

4. Computational Aspects

• No algorithm complexity analysis provided
• For large n, how can δ(n) be computed efficiently?
• Numerical data only extends to n=127; what is the behavior for larger n?

Impact Assessment

Contribution to the Field

• Combinatorics: Provides new tools for studying modular properties of partition dimensions (Od function, parent partition theory)
• Representation Theory: Provides combinatorial foundation for refined classification of spinor representations of symmetric groups
• Algorithms: Explicit formulas for sparse numbers enable fast computation

Practical Value

• Sparse numbers appear frequently in computer science (e.g., Fibonacci coding); formulas have potential applications
• Provides clear technical framework for further research

Reproducibility

• Definitions and proofs are complete and theoretically verifiable
• Appendix data can be used for independent verification
• No code provided, but algorithm descriptions are sufficient for implementation

Applicable Scenarios

1. Theoretical Research:
   • Study of modular properties of symmetric group representations
   • Exploration of fine structure of partition statistics
   • Development of new directions in p-core theory
2. Computational Applications:
   • Fast determination of partition dimension distribution for sparse numbers
   • Optimization in symmetric group representation computation
3. Educational Use:
   • Illustration of connections between combinatorics and algebraic representation theory
   • Case study of recursive methods and generating function techniques

In-Depth Analysis of Technical Highlights

Power of the Od Function

Though seemingly simple, the Od function is cleverly designed:

1. Multiplicative property enables decomposed computation: $\text{Od}(n!) = \prod_{r=1}^n \text{Od}(r)$
2. Lemma 33's formula reduces Od values of factorials to binary statistics D(n) and ν(n)
3. Coordination with hook-length formula: Product terms in Frobenius formula (Proposition 18) can be processed term-by-term via Od

Geometric Intuition of Workhorse Formula

Proposition 36 can be understood as:

• Adding a $2^R$-hook from μ to λ corresponds to a "jump" in the Ferrers diagram
• The $s_2$ term records binary prefix information of n and $h_\mu^\lambda$
• The $\eta_\mu^\lambda$ term precisely quantifies sign changes from the "jump" interacting with other hooks
• When $m < 2^{R-1}$, the geometric structure is simple, $s_2$ terms cancel, leaving only η

Combinatorics of Parent Partition Counting

Analysis of Type II parent partitions (Lemmas 58, 61) reveals deep combinatorial insight:

• Partitioning $\{1,\ldots,2^R\}$ into $P_2^\uparrow$ and $P_2^\downarrow$ reflects the critical role of $2^{R-1}$
• Introduction of parity gap statistic G captures global parity information
• Lemma 59 connects G to partition length and size parity, revealing underlying structure

Comparison with Related Work

This paper's unique contributions:

• Not only counts modulo 4 partitions, but distinguishes between the two odd residue classes
• Introduces new tools (Od function, workhorse formula) rather than merely applying existing theory
• Provides explicit formulas for specific n rather than only asymptotic or recursive results

Difficulty Analysis of Open Problems

The author analyzes in Section 6 the difficulty of the case $2^{R-1} < m < 2^R$:

Core Obstacle: $S_{P(\mu)}(\mu)$ no longer depends only on |μ|, but on the fine structure of the hook length set H(μ).

Concrete Example (provided by author):

• $\mu = (17,3,1) \vdash 21$: $S_{P(\mu)}(\mu) = 8$
• $\nu = (12,3,3,2,1) \vdash 21$: $S_{P(\nu)}(\nu) = 0$

This shows that new invariants are needed to classify odd partitions of 21.

Possible Directions:

1. Search for refined invariants of H(μ) (e.g., some kind of "modulo 4 signature")
2. Consider multi-level core structures ($2^R$-core of $2^{R-1}$-core, etc.)
3. Use computer algebra systems to explore patterns for small n, seeking conjectures

Concluding Remarks

This paper is an excellent work at the intersection of combinatorics and algebraic representation theory. Through introducing new tools (Od function, parent partition theory) and establishing core techniques (workhorse formula), it partially resolves the refined problem of counting odd-dimensional partitions modulo 4. The complete resolution of the sparse number case demonstrates the effectiveness of the method, while the difficulty of unresolved cases is honestly presented, pointing directions for future research.

The paper's main value lies in:

1. Methodological Contribution: Provides a systematic framework for handling modular properties of partition dimensions
2. Concrete Results: Gives elegant formulas for special cases like sparse numbers
3. Problem-Oriented: Clearly delineates resolved and unresolved boundaries

For researchers in combinatorics and representation theory, this paper is an important reference for deeply understanding modular properties of partitions; for students wishing to learn advanced combinatorial techniques, it demonstrates the elegant combination of recursive methods, generating functions, and binary analysis.

Selected References

1 I. G. Macdonald. "On the Degrees of the Irreducible Representations of Symmetric Groups". Bulletin of the London Mathematical Society, 1971. (Foundational work on enumeration of odd partitions)

5 A. Ayyer, A. Prasad, S. Spallone. "Odd partitions in Young's lattice". arXiv:1601.01776, 2016. (Structural properties of odd partitions)

13 K. Davis, W. Webb. "Pascal's Triangle modulo 4". Fibonacci Quarterly, 1989. (Source of D(n) statistic)

Suggested Ratings:

• Originality: 8/10 (New tools and methods, though the problem itself is specialized)
• Technical Depth: 9/10 (Rigorous proofs, complex techniques)
• Completeness: 7/10 (Only resolves partial cases)
• Impact: 7/10 (Important for specific field, but narrow application scope)
• Readability: 8/10 (Clear writing, abundant examples)

Overall Assessment: This is a high-quality combinatorics paper that achieves substantial progress on a limited but important problem, laying solid foundations for further research.