Recently, the sparse representation of multivariate data has gained great popularity in real-world applications like neural activity analysis. Many previous analyses for these data utilize sparse principal component a...
详细信息
Recently, the sparse representation of multivariate data has gained great popularity in real-world applications like neural activity analysis. Many previous analyses for these data utilize sparse principal component analysis (SPCA) to obtain a sparse representation. However, l(0)-norm based SPCA suffers from non-differentiability and local optimum problems due to non-convex regularization. Additionally, extracting dependencies between task parameters and feature responses is essential for further analysis while SPCA usually generates components without demixing these dependencies. To address these problems, we propose a novel approach, demixed sparse principal component analysis (dSPCA), that relaxes the non-convex constraints into convex regularizers, e.g., l(1)-norm and nuclear norm, and demixes dependencies of feature response on various task parameters by optimizing the loss function with marginalized data. The sparse and demixed components greatly improve the interpretability of the multivariate data. We also develop a parallelproximal algorithm to accelerate the optimization for hybrid regularizers based on our method. We provide theoretical analyses for error bound and convergency. We apply our method on simulated datasets to evaluate its time cost, the ability to explain the demixed information, and the ability to recover sparsity for the reconstructed data. Finally, we successfully separate the neural activity into different task parameters like stimulus or decision, and visualize the demixed components based on the real-world dataset.
proximal splitting algorithms play a central role in finding the numerical solution of convex optimization problems. This paper addresses the problem of stereo matching of multi-component images by jointly estimating ...
详细信息
proximal splitting algorithms play a central role in finding the numerical solution of convex optimization problems. This paper addresses the problem of stereo matching of multi-component images by jointly estimating the disparity and the illumination variation. The global formulation being non-convex, the problem is addressed by solving a sequence of convex relaxations. Each convex relaxation is non trivial and involves many constraints aiming at imposing some regularity on the solution. Experiments demonstrate that the method is efficient and provides better results compared with other approaches.
暂无评论