arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.23812cs.LGmath.PRstat.ML

迭代原子细化:字典学习中的单调性原理

Iterative Atom Refinement: A Monotonicity Principle for Dictionary Learning

  • Stanford University(斯坦福大学)
  • Universidad Carlos III de Madrid(马德里卡洛斯三世大学)
  • Pennsylvania State University(宾夕法尼亚州立大学)
  • University of California, Merced(加州大学默塞德分校)

机构由 AI 辅助整理,请以论文原文为准。

Alexander Christie, Miguel Moscoso, Alexei Novikov, George Papanicolaou, Chrysoula Tsogka

AI总结:

本文提出迭代原子细化(IAR)算法,通过迭代选择强相关观测并平均更新方向,借助单调性原理证明三步内可恢复字典原子,实验验证了理论。

AI中文摘要:

字典学习旨在从观测数据 ${\f y}_i = A{\f x}_i$ 中恢复未知字典 $A$,其中系数向量 ${\f x}_i$ 是稀疏的。我们提出了迭代原子细化(IAR)算法,这是一种恢复单个字典原子的简单方法。从随机方向开始,IAR 反复选择与当前迭代向量相关性最强的观测数据,并通过平均所选数据来更新方向。我们的主要贡献是建立了 IAR 的严格收敛理论。利用高维概率估计和一种新颖的原子选择概率单调性原理,我们证明了一个原子的微小初始优势会被放大,直到该原子被分离出来。在我们的模型假设下,IAR 仅需三步细化即可识别出一个生成原子。数值实验支持该理论,并表明所得动力学能够准确捕捉字典细化中观察到的行为。

英文摘要:

Dictionary learning seeks to recover an unknown dictionary $A$ from observations ${\bf y}_i = A{\bf x}_i$ with sparse coefficient vectors ${\bf x}_i$. We introduce the \emph{Iterative Atom Refinement} (IAR) algorithm, a simple procedure for recovering individual dictionary atoms. Starting from a random direction, IAR repeatedly selects the observations most strongly correlated with the current iterate and updates the direction by averaging the selected data. Our main contribution is a rigorous convergence theory of IAR. Using high-dimensional probabilistic estimates and a novel monotonicity principle for atom-selection probabilities, we show that a small initial advantage of one atom is amplified until that atom is isolated. Under our model assumptions, IAR identifies a generating atom after only three refinement steps. Numerical experiments support the theory and show that the resulting dynamics accurately capture the behavior observed in dictionary refinement.

↑