SearcharxivSearch

arXiv subjects

Zihang Fang

Publications and source records attributed to Zihang Fang.

3 recordsLinked to original sources

Self-Referential Leading Digits of Exponential Sequences: Arithmetic Structure and Certified Search

For $c>1$ and an integer radix $b\ge2$, we study the positive integers $m$ for which $mb^k\le c^m<(m+1)b^k$ for some $k\ge0$; for integer $c$, this is the self-prefix leading-digit condition. We derive an exact shrinking-target criterion; for $c\ge2$, an exact signed-discrepancy identity isolates both infinitude and the conjectural logarithmic count. For $c\ge2$ with nonintegral logarithmic slope, Lambert $W_{-1}$ inversion produces a candidate sequence with an eventual two-gap law and an exact counting formula; for $(c,b)=(2,10)$ all consecutive candidate gaps are $3$ or $4$. For algebraic $c$ with irrational $\log_b c$, the Lambert-root phases satisfy deterministic moving-target asymptotics in an explicit nontrivial power range strictly below the critical scale. For irrational logarithmic slope, actual hits obey fixed-difference and arithmetic-chain rigidity; for multiplicatively independent integer parameters, coherent endpoint hits at floor resonance centers force every intermediate term. Finally, set $\rho=\{\log_b c\}$. For fixed multiplicatively independent integers $c,b$, an interpolated continued-fraction locator has bit complexity $O(N^{1-1/\nu}\operatorname{polylog}N)$ for every $\nu>\mu(\rho)$. We give an explicit certified instance for $(2,10)$, whose infinitude remains open.

math.NT

Bundle Adjustment in the Eager Mode

Bundle adjustment (BA) is a critical technique in various robotic applications such as simultaneous localization and mapping (SLAM), augmented reality (AR), and photogrammetry. BA optimizes parameters such as camera poses and 3D landmarks to align them with observations. With the growing importance of deep learning in perception systems, there is an increasing need to integrate BA with deep learning frameworks for enhanced reliability and performance. However, widely-used C++-based BA libraries, such as GTSAM, g$^2$o, and Ceres Solver, lack native integration with modern deep learning libraries like PyTorch. This limitation affects their flexibility, ease of debugging, and overall implementation efficiency. To address this gap, we introduce an eager-mode BA library seamlessly integrated with PyTorch with high efficiency. Our approach includes a sparsity-aware auto-differentiation design and GPU-accelerated sparse operations designed for 2nd-order optimization. Our eager-mode BA on GPU demonstrates substantial runtime efficiency, achieving an average speedup of 18.5$\times$, 22$\times$, and 23$\times$ across all benchmarks compared to GTSAM, g$^2$o, and Ceres, respectively.

cs.RO

PyPose v0.6: The Imperative Programming Interface for Robotics

PyPose is an open-source library for robot learning. It combines a learning-based approach with physics-based optimization, which enables seamless end-to-end robot learning. It has been used in many tasks due to its meticulously designed application programming interface (API) and efficient implementation. From its initial launch in early 2022, PyPose has experienced significant enhancements, incorporating a wide variety of new features into its platform. To satisfy the growing demand for understanding and utilizing the library and reduce the learning curve of new users, we present the fundamental design principle of the imperative programming interface, and showcase the flexible usage of diverse functionalities and modules using an extremely simple Dubins car example. We also demonstrate that the PyPose can be easily used to navigate a real quadruped robot with a few lines of code.

cs.RO