提出新模型，让大规模数据聚类速度提升且不丢精度
Scalable Deep Subspace Clustering Network
- 用关键点近似替代全量相似度计算，降低复杂度
- 在10万样本上比现有方法快10倍以上，精度接近顶尖水平
- 适合处理海量高维数据的聚类任务
子空间聚类方法因构建 $n\times n$ 相似度矩阵和进行谱分解的 $O(n^3)$ 复杂度而面临固有的可扩展性瓶颈。尽管基于深度学习的方法提升了特征提取能力，但仍通过穷举成对相似性计算维持这一计算瓶颈。本文提出 SDSNet（Scalable Deep Subspace Network），一种实现 $\mathcal{O}(n)$ 复杂度的深度子空间聚类框架，其核心包括：(1) 基于关键点的近似方法，避免构建完整亲缘矩阵；(2) 联合优化自编码器重建与自表达目标；(3) 在因子化表示上直接执行谱聚类。该框架结合卷积自编码器与保持子空间结构的约束。实验表明，SDSNet 在保持与前沿方法相当聚类质量的同时，显著提升了计算效率。
原文摘要 · Abstract (English)
Subspace clustering methods face inherent scalability limits due to the $O(n^3)$ cost (with $n$ denoting the number of data samples) of constructing full $n\times n$ affinities and performing spectral decomposition. While deep learning-based approaches improve feature extraction, they maintain this computational bottleneck through exhaustive pairwise similarity computations. We propose SDSNet (Scalable Deep Subspace Network), a deep subspace clustering framework that achieves $\mathcal{O}(n)$ complexity through (1) landmark-based approximation, avoiding full affinity matrices, (2) joint optimization of auto-encoder reconstruction with self-expression objectives, and (3) direct spectral clustering on factorized representations. The framework combines convolutional auto-encoders with subspace-preserving constraints. Experimental results demonstrate that SDSNet achieves comparable clustering quality to state-of-the-art methods with significantly improved computational efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。