An official website of the United States government
Abstract It is increasingly interesting to model the relationship between two sets of high-dimensional measurements with potentially high correlations. Canonical correlation analysis (CCA) is a classical tool that explores the dependency of two multivariate random variables and extracts canonical pairs of highly correlated linear combinations. Driven by applications in genomics, text mining, and imaging research, among others, many recent studies generalize CCA to high-dimensional settings. However, most of them either rely on strong assumptions on covariance matrices, or do not produce nested solutions. We propose a new sparse CCA (SCCA) method that recasts high-dimensional CCA as an iterative penalized least squares problem. Thanks to the new iterative penalized least squares formulation, our method directly estimates the sparse CCA directions with efficient algorithms. Therefore, in contrast to some existing methods, the new SCCA does not impose any sparsity assumptions on the covariance matrices. The proposed SCCA is also very flexible in the sense that it can be easily combined with properly chosen penalty functions to perform structured variable selection and incorporate prior information. Moreover, our proposal of SCCA produces nested solutions and thus provides great convenient in practice. Theoretical results show that SCCA can consistently estimate the true canonical pairs with an overwhelming probability in ultra-high dimensions. Numerical results also demonstrate the competitive performance of SCCA.
more »
« less
Efficient Globally Convergent Stochastic Optimization for Canonical Correlation Analysis
We study the stochastic optimization of canonical correlation analysis (CCA), whose objective is nonconvex and does not decouple over training samples. Although several stochastic gradient based optimization algorithms have been recently proposed to solve this problem, no global convergence guarantee was provided by any of them. Inspired by the alternating least squares/power iterations formulation of CCA, and the shift-and-invert preconditioning method for PCA, we propose two globally convergent meta-algorithms for CCA, both of which transform the original problem into sequences of least squares problems that need only be solved approximately. We instantiate the meta-algorithms with state-of-the-art SGD methods and obtain time complexities that significantly improve upon that of previous work. Experimental results demonstrate their superior performance.
more »
« less
- Award ID(s):
- 1302662
- PAR ID:
- 10025958
- Date Published:
- Journal Name:
- Advances in Neural Information Processing Systems 29 (NIPS 2016)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Full waveform inversion (FWI) and least-squares reverse time migration (LSRTM) are popular imaging techniques that can be solved as PDE-constrained optimization problems. Due to the large-scale nature, gradient- and Hessian-based optimization algorithms are preferred in practice to find the optimizer iteratively. However, a balance between the evaluation cost and the rate of convergence needs to be considered. We propose the use of Anderson acceleration (AA), a popular strategy to speed up the convergence of fixed-point iterations, to accelerate a gradient descent method. We show that AA can achieve fast convergence that provides competitive results with some quasi-Newton methods. Independent of the dimensionality of the unknown parameters, the computational cost of implementing the method can be reduced to an extremely lowdimensional least-squares problem, which makes AA an attractive method for seismic inversion.more » « less
-
This paper tackles a problem in line-assisted VO/VSLAM: accurately solving the least squares pose optimization with unreliable 3D line input. The solution we present is good line cutting, which extracts the most-informative sub-segment from each 3D line for use within the pose optimization formulation. By studying the impact of line cutting towards the information gain of pose estimation in line-based least squares problem, we demonstrate the applicability of improving pose estimation accuracy with good line cutting. To that end, we describe an efficient algorithm that approximately approaches the joint optimization problem of good line cutting. The proposed algorithm is integrated into a state-of-the-art line-assisted VSLAM system. When evaluated in two target scenarios of line-assisted VO/VSLAM, low-texture and motion blur, the accuracy of pose tracking is improved, while the robustness is preserved.more » « less
-
Real-world problems often have parameters that are uncertain during the optimization phase; stochastic optimization or stochastic programming is a key approach introduced by Beale and by Dantzig in the 1950s to address such uncertainty. Matching is a classical problem in combinatorial optimization. Modern stochastic versions of this problem model problems in kidney exchange, for instance. We improve upon the current-best approximation bound of 3.709 for stochastic matching due to Adamczyk et al. (in: Algorithms-ESA 2015, Springer, Berlin, 2015) to 3.224; we also present improvements on Bansal et al. (Algorithmica 63(4):733–762, 2012) for hypergraph matching and for relaxed versions of the problem. These results are obtained by improved analyses and/or algorithms for rounding linear-programming relaxations of these problems.more » « less
-
We develop algorithms to automate discovery of stochastic dynamical system models from noisy, vector-valued time series. By discovery, we mean learning both a nonlinear drift vector field and a diagonal diffusion matrix for an Itô stochastic differential equation in Rd . We parameterize the vector field using tensor products of Hermite polynomials, enabling the model to capture highly nonlinear and/or coupled dynamics. We solve the resulting estimation problem using expectation maximization (EM). This involves two steps. We augment the data via diffusion bridge sampling, with the goal of producing time series observed at a higher frequency than the original data. With this augmented data, the resulting expected log likelihood maximization problem reduces to a least squares problem. We provide an open-source implementation of this algorithm. Through experiments on systems with dimensions one through eight, we show that this EM approach enables accurate estimation for multiple time series with possibly irregular observation times. We study how the EM method performs as a function of the amount of data augmentation, as well as the volume and noisiness of the data.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
