Provable Subspace Clustering: When LRR meets SSC
نویسندگان
چکیده
• Random Sampling: X are iid uniform from unit sphere in S`. • Random Subspace: S` are spanned by d iid uniform vectors in R. GRAPH CONNECTIVITY What about the second design objective? • Nashihatkon & Hartley nailed that connectivity is NOT a generic property for SSC in general when d ≥ 4. • What about for LRR? Assume random sampling and random subspace, we have: Proposition 1 Under independent subspace assumption, solution to LRR is class-wise dense, namely each diagonal block of the matrix C is all non-zero.
منابع مشابه
Supplementary material for “ Provable Subspace Clustering : When LRR meets SSC ”
The supplementary material is organized as follows. In Section A and B, we provide the detailed proof of respectively the deterministic and randomized guarantee for LRSSC. In Section C, we derive the fast Alternating Direction Methods of Multipliers (ADMM) algorithm for LRSSC and NoisyLRSSC and verify its convergence guarantee. In Section D, additional numerical experiments of LRSSC are provide...
متن کاملA Self-Training Subspace Clustering
Accurate identification of the cancer types is essential to cancer diagnoses and treatments. Since cancer tissue and normal tissue have different gene expression, gene expression data can be used as an efficient feature source for cancer classification. However, accurate cancer classification directly using original gene expression profiles remains challenging due to the intrinsic high-dimensio...
متن کاملEfficient Subspace Segmentation via Quadratic Programming
We explore in this paper efficient algorithmic solutions to robust subspace segmentation. We propose the SSQP, namely Subspace Segmentation via Quadratic Programming, to partition data drawn from multiple subspaces into multiple clusters. The basic idea of SSQP is to express each datum as the linear combination of other data regularized by an overall term targeting zero reconstruction coefficie...
متن کاملSegmentation of Subspaces in Sequential Data
We propose Ordered Subspace Clustering (OSC) to segment data drawn from a sequentially ordered union of subspaces. Similar to Sparse Subspace Clustering (SSC) we formulate the problem as one of finding a sparse representation but include an additional penalty term to take care of sequential data. We test our method on data drawn from infrared hyper spectral, video and motion capture data. Exper...
متن کاملA Review on Fixed-Rank Representation for Supervised Learning Using Neural Network
This study focused on development and application of efficient algorithm for clustering and classification of supervised visual data. In machine learning clustering classification and clustering is most useful techniques in pattern recognition and computer vision. Existing techniques for subspace clustering makes use of rank minimization and spars based that are computationally expensive and ma...
متن کامل