Elastic Net subspace clustering applied to pop/rock music structure analysis

Panagakis, Yannis; Kotropoulos, Constantine

doi:10.1016/j.patrec.2013.10.021

Cited by 27 publications

(19 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the elastic net optimization problem [50] has been recently introduced for subspace clustering in [12,13,32], such prior work does not provide an efficient algorithm that can handle large-scale datasets. In fact, such prior work solves the elastic net problem using existing algorithms that require calculations involving the full data matrix A (e.g., the accelerated proximal gradient (APG) [2] is used in [12] and the linearized alternating direction method (LADM) [25] is used in [32]). Here, we propose to solve the elastic net problem (4) with an active-set algorithm that is more efficient than both APG and LADM, and can handle largescale datasets.…”

Section: A New Active-set Algorithmmentioning

confidence: 99%

“…To bridge the gap between the subspace preserving and connectedness properties, [45,32,13] propose to use mixed norms. For example, the low rank sparse subspace clustering (LRSSC) method [45], which uses a mixed ℓ 1 and nuclear norm regularizer, is shown to give a subspace preserving representation under conditions which are similar to but stronger than those of SSC.…”

Section: Introductionmentioning

confidence: 99%

“…However, no theoretical justification is provided in [28,22] for the benefit of their methods. Another issue with the aforementioned methods [45,32,13,28,22] is that they do not provide efficient algorithms to deal with large-scale datasets. To address this issue, [5] proposes to find the representation of X by a few anchor points that are sampled from X and then perform spectral clustering on the anchor graph.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering

You

Robinson

et al. 2016

2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

210

169

View full text Add to dashboard Cite

show abstract

Section: A New Active-set Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering

You

Robinson

et al. 2016

2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

210

169

View full text Add to dashboard Cite

show abstract

“…Mixture of 1 and 2 norms was used in [25] to take advantage of subspace preserving of the 1 norm and the dense connectivity of the 2 norm. Later in [39], an oracle-based algorithm, dubbed ORacle Guided Elastic Net solver (ORGEN), was proposed to identify a support set for each sample efficiently.…”

Section: Related Workmentioning

confidence: 99%

Scalable and robust sparse subspace clustering using randomized clustering and multilayer graphs

2019

View full text Add to dashboard Cite

Sparse subspace clustering (SSC) is one of the current state-of-the-art methods for partitioning data points into the union of subspaces, with strong theoretical guarantees. However, it is not practical for large data sets as it requires solving a LASSO problem for each data point, where the number of variables in each LASSO problem is the number of data points. To improve the scalability of SSC, we propose to select a few sets of anchor points using a randomized hierarchical clustering method, and, for each set of anchor points, solve the LASSO problems for each data point allowing only anchor points to have a non-zero weight (this reduces drastically the number of variables). This generates a multilayer graph where each layer corresponds to a different set of anchor points. Using the Grassmann manifold of orthogonal matrices, the shared connectivity among the layers is summarized within a single subspace. Finally, we use k-means clustering within that subspace to cluster the data points, similarly as done by spectral clustering in SSC. We show on both synthetic and real-world data sets that the proposed method not only allows SSC to scale to large-scale data sets, but that it is also much more robust as it performs significantly better on noisy data and on data with close susbspaces and outliers, while it is not prone to oversegmentation.

show abstract

“…For instance, Zheng and Liu [7] have developed a Lasso operator to identify the most informative features for cancer classification, where Lasso enforces automatic feature selection by forcing at least some features to zero. Panagakis et al [34] have developed a new similarity measure based on the matrix Elastic Net regularization to efficiently deal with highly correlated audio feature vectors. Marafino et al [35] have proposed an efficient sparse feature selection method for biomedical text classification using the Elastic Net.…”

Section: Introductionmentioning

confidence: 99%

Identifying the most informative features using a structurally interacting elastic net

et al. 2019

View full text Add to dashboard Cite

Feature selection can efficiently identify the most informative features with respect to the target feature used in training. However, state-of-the-art vector-based methods are unable to encapsulate the relationships between feature samples into the feature selection process, thus leading to significant information loss. To address this problem, we propose a new graph-based structurally interacting elastic net method for feature selection. Specifically, we commence by constructing feature graphs that can incorporate pairwise relationship between samples. With the feature graphs to hand, we propose a new information theoretic criterion to measure the joint relevance of different pairwise feature combinations with respect to the target feature graph representation. This measure is used to obtain a structural interaction matrix where the elements represent the proposed information theoretic measure between feature pairs. We then formulate a new optimization model through the combination of the structural interaction matrix and an elastic net regression model for the feature subset selection problem. This allows us to a) preserve the information of the original vectorial space, b) remedy the information loss of the original feature space caused by using graph representation, and c) promote a sparse solution and also encourage correlated features to be selected. Because the proposed optimization problem is non-convex, we develop an efficient alternating direction multiplier method (ADMM) to locate the optimal solutions. Extensive experiments on various datasets demonstrate the effectiveness of the proposed method.

show abstract

Elastic Net subspace clustering applied to pop/rock music structure analysis

Cited by 27 publications

References 19 publications

Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering

Oracle Based Active Set Algorithm for Scalable Elastic Net Subspace Clustering

Scalable and robust sparse subspace clustering using randomized clustering and multilayer graphs

Identifying the most informative features using a structurally interacting elastic net

Contact Info

Product

Resources

About