Searcharxiv⌕ Search

arXiv subjects

Frederick Rajasekaran

Publications and source records attributed to Frederick Rajasekaran.

5 recordsLinked to original sources

Definition Grounding in Theory-Level Auto-formalization

LLM-based auto-formalization systems can now generate formally verified Lean code at unprecedented rates. However, a formalization advances mathematical research only when further mathematical results can be built on it, especially when its definitions and helper lemmas are general enough to be imported and used idiomatically. We propose definition grounding, where definition formalizations are continuously judged and revised, as a method to ensure formalization quality. We find that agents formalizing definitions from textbook text often produce definitions that later statements cannot use, and only 25%-50% of their definition sets are compatible with all later statements. However, clarifying the informal definitions can improve this rate up to 77.5% averaged over all models. Furthermore, stating a small set of basic properties without proving them can improve model judges' capability to detect definitions with compatibility issues, at only about 11% of the definition formalization cost. These results are verified with ReuseBench, a benchmark we curated for formalizing new mathematical fields and theories.

cs.AI↗

Gaussian limits of lattice Higgs models with complete symmetry breaking

Given any compact connected matrix Lie group $G$ and any lattice dimension $d\ge 2$, we construct a massive Gaussian scaling limit for the $G$-valued lattice Yang-Mills-Higgs theory in the "complete breakdown of symmetry" regime. This limit arises as the lattice spacing tends to zero and the (inverse) gauge coupling constant tends to infinity sufficiently fast, causing the theory to "abelianize" and yield a Gaussian limit. This complements a recent work by Chatterjee (arXiv:2401.10507), which obtained a similar scaling limit in the special case $G= SU(2)$.

math.PR↗

Fluctuations of Eigenvalues for Generalized Patterned Gaussian Random Matrices

In this work, we study a class of random matrices which interpolate between the Wigner matrix model and various types of patterned random matrices such as random Toeplitz, Hankel, and circulant matrices. The interpolation mechanism is through the correlations of the entries, and thus these interpolating models are highly inhomogeneous in their correlation structure. Historically, the study of random matrices has focused on homogeneous models, i.e., those with imposed structure (such as independence of the entries), as such restrictions significantly simplify computations related to the models. However, in this paper we demonstrate that for these interpolating inhomogenous models, the fluctuations of the linear eigenvalue statistics are approximately Gaussian. To handle the difficulties that come with inhomogeneity in the entries, we incorporate combinatorial arguments and recent tools from non-asymptotic random matrix theory.

math.PR↗

Numerical Evidence of Exponential Mixing by Alternating Shear Flows

We performed a numerical study of the efficiency of mixing by alternating horizontal and vertical shear ``wedge'' flows on the two-dimensional torus. Our results suggest that except in cases where each individual flow is applied for only a short time, these flows produce exponentially fast mixing. The observed mixing rates are higher when the individual flow times are shorter (but not too short), and randomizing either the flow times or phase shifts of the flows does not appear to enhance mixing (again when the flow times are not too short). In fact, the latter surprisingly seems to inhibit it slightly.

math.AP↗

Set Partition Patterns and the Dimension Index

The notion of containment and avoidance provides a natural partial ordering on set partitions. Work of Sagan and of Goyt has led to enumerative results in avoidance classes of set partitions, which were refined by Dahlberg et al. through the use of combinatorial statistics. We continue this work by computing the distribution of the dimension index (a statistic arising from the supercharacter theory of finite groups) across certain avoidance classes of partitions. In doing so we obtain a novel connection between noncrossing partitions and 321-avoiding permutations, as well as connections to many other combinatorial objects such as Motzkin and Fibonacci polynomials.

math.CO↗