发表机构
University of Alberta; Alberta Machine Intelligence Institute (Amii)(阿尔伯塔大学; 阿尔伯塔机器智能研究所(Amii))
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
该研究针对现有解耦表示方法未考虑因子空间非均匀几何的问题,提出FactoMap方法,通过引入因子空间结构学习可解释原型,实验证明其能保持因子连续性并实现因子解耦。
AI 中文摘要
许多解耦表示方法用欧几里得乘积坐标表示生成因子,尽管底层因子空间可能存在缠绕、坍缩或位置依赖的几何结构。我们引入因子空间结构,结合因子域、生成器诱导的识别关系以及位置依赖的尺度,以区分具有不同因子几何的拓扑等价空间。我们表明统计独立的因子不必在几何上可分离:色调和尺度产生的影响以不同速率增长,产生各向异性,无法通过固定重缩放消除。我们提出因子空间地形地图(Factor-Space Topographic Map,FactoMap),它学习由因子空间格点索引的可解释原型。地形学习将格点的周期性、坍缩和非均匀范围传递到表示中。实验表明,匹配该结构可保持因子连续性,并实现底层因子的解耦。
英文摘要
Many disentanglement methods represent generative factors using Euclidean product coordinates, although the underlying factor spaces may wrap, collapse, or have position-dependent geometry. We introduce factor-space structure, combining factor domains, generator-induced identifications, and position-dependent scales to distinguish topologically equivalent spaces with different factor geometries. We show that statistically independent factors need not be geometrically separable: hue and scale produce effects that grow at different rates, yielding anisotropy that no fixed rescaling removes. We propose the Factor-Space Topographic Map (FactoMap), which learns interpretable prototypes indexed by a factor-space lattice. Topographic learning transfers the lattice's periodicity, collapses, and non-uniform extent to the representation. Experiments show that matching this structure preserves factor continuity and enables disentanglement of the underlying factors.