arXiv subjects

Chen-Yang Su

Publications and source records attributed to Chen-Yang Su.

4 recordsLinked to original sources

arXiv2026

Parameterized partial theta identities and a unified $q$-difference proof

We establish five families of integer-parameter extensions of Ramanujan's partial theta identities. The families follow from a common two-parameter specialization of Andrews' transformation, for which we give an independent proof based on a $q$-difference recurrence and a boundary estimate. Specializations of the integer parameter recover six identities from Ramanujan's lost notebook. As an application, a residue argument applied to the fifth family gives an integer-parameter extension of Lovejoy's residual identity, from which we construct a corresponding family of conjugate Bailey pairs.

math.NT↗

arXiv2026

Parallel packing a square with isosceles right triangles and equilateral triangles

Suppose that $I$ is a unit square. Let $T$ (resp. $\Delta$) be an isosceles right triangle (resp. an equilateral triangle). We prove that any collection of triangles homothetic to $T$ (resp. $\Delta$), whose total area does not exceed $\frac{1}{2}$ (resp. $\frac{\sqrt{3}}{4}$), can be parallel packed into $I$. These upper bounds are tight.

math.CO↗

arXiv2020

Regularized Inverse Reinforcement Learning

Inverse Reinforcement Learning (IRL) aims to facilitate a learner's ability to imitate expert behavior by acquiring reward functions that explain the expert's decisions. Regularized IRL applies strongly convex regularizers to the learner's policy in order to avoid the expert's behavior being rationalized by arbitrary constant rewards, also known as degenerate solutions. We propose tractable solutions, and practical methods to obtain them, for regularized IRL. Current methods are restricted to the maximum-entropy IRL framework, limiting them to Shannon-entropy regularizers, as well as proposing the solutions that are intractable in practice. We present theoretical backing for our proposed IRL method's applicability for both discrete and continuous controls, empirically validating our performance on a variety of tasks.

cs.LG↗

arXiv2018

Four identities related to third order mock theta functions

Ramanujan presented four identities for third order mock theta functions in his Lost Notebook. In 2005, with the aid of complex analysis, Yesilyurt first proved these four identities. Recently, Andrews et al. provided different proofs by using $q$-series. In this paper, in view of some identities of a universal mock theta function \begin{align*} g(x;q)=x^{-1}\left(-1+\sum_{n=0}^{\infty}\frac{q^{n^{2}}}{(x;q)_{n+1}(qx^{-1};q)_{n}}\right), \end{align*} we establish new proofs of these four identities. In particular, by means of an identity of $g(x;q)$ given by Ramanujan and some theta function identities due to Mortenson, we find a new simple proof of the fourth identity.

math.CO↗