FAASTA: A fast solver for total-variation regularization of ill-conditioned problems with application to brain imaging

The total variation (TV) penalty, as many other analysis-sparsity problems, does not lead to separable factors or a proximal operatorwith a closed-form expression, such as soft thresholding for the $\ell\_1$ penalty. As a result, in a variational formulation of an inverse problem or statisticallearning estimation, it leads to challenging non-smooth optimization problemsthat are often solved with elaborate single-step first-order methods. When thedata-fit term arises from empirical measurements, as in brain imaging, it isoften very ill-conditioned and without simple structure. In this situation, in proximal splitting methods, the computation cost of thegradient step can easily dominate each iteration. Thus it is beneficialto minimize the number of gradient steps.We present fAASTA, a variant of FISTA, that relies on an internal solver forthe TV proximal operator, and refines its tolerance to balance computationalcost of the gradient and the proximal steps. We give benchmarks andillustrations on "brain decoding": recovering brain maps from noisymeasurements to predict observed behavior. The algorithm as well as theempirical study of convergence speed are valuable for any non-exact proximaloperator, in particular analysis-sparsity problems.

[1]  Marc Teboulle,et al.  A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems , 2009, SIAM J. Imaging Sci..

[2]  Mohamed-Jalal Fadili,et al.  Low Complexity Regularization of Linear Inverse Problems , 2014, ArXiv.

[3]  Gaël Varoquaux,et al.  Total Variation Regularization for fMRI-Based Prediction of Behavior , 2011, IEEE Transactions on Medical Imaging.

[4]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[5]  Patrick L. Combettes,et al.  Proximal Splitting Methods in Signal Processing , 2009, Fixed-Point Algorithms for Inverse Problems in Science and Engineering.

[6]  E. Sidky,et al.  Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization , 2008, Physics in medicine and biology.

[7]  Vincent Frouin,et al.  Predictive support recovery with TV-Elastic Net penalty and logistic regression: An application to structural MRI , 2014, 2014 International Workshop on Pattern Recognition in Neuroimaging.

[8]  Gaël Varoquaux,et al.  Identifying Predictive Regions from fMRI with TV-L1 Prior , 2013, 2013 International Workshop on Pattern Recognition in Neuroimaging.

[9]  Mark W. Schmidt,et al.  Convergence Rates of Inexact Proximal-Gradient Methods for Convex Optimization , 2011, NIPS.

[10]  Gaël Varoquaux,et al.  Benchmarking solvers for TV-ℓ1 least-squares and logistic regression in brain imaging , 2014, 2014 International Workshop on Pattern Recognition in Neuroimaging.

[11]  Marc Teboulle,et al.  Fast Gradient-Based Algorithms for Constrained Total Variation Image Denoising and Deblurring Problems , 2009, IEEE Transactions on Image Processing.

[12]  Antonin Chambolle,et al.  A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging , 2011, Journal of Mathematical Imaging and Vision.

[13]  I. Daubechies,et al.  An iterative thresholding algorithm for linear inverse problems with a sparsity constraint , 2003, math/0307152.

[14]  A. Ishai,et al.  Distributed and Overlapping Representations of Faces and Objects in Ventral Temporal Cortex , 2001, Science.

[15]  Michael Eickenberg,et al.  Machine learning for neuroimaging with scikit-learn , 2014, Front. Neuroinform..

[16]  Stephen P. Boyd,et al.  Proximal Algorithms , 2013, Found. Trends Optim..

[17]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[18]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[19]  Jorge Nocedal,et al.  On the limited memory BFGS method for large scale optimization , 1989, Math. Program..