1. |
- Kronvall, Ted, et al.
(författare)
-
Hyperparameter-free sparse regression of grouped variables
- 2017
-
Ingår i: Conference Record of the 50th Asilomar Conference on Signals, Systems and Computers, ACSSC 2016. - 9781538639542 ; , s. 394-398
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we introduce a novel framework for semi-parametric estimation of an unknown number of signals, each parametrized by a group of components. Via a reformulation of the covariance fitting criteria, we formulate a convex optimization problem over a grid of candidate representations, promoting solutions with only a few active groups. Utilizing the covariance fitting allows for a hyperparameter-free estimation procedure, highly robust against coherency between candidates, while still allowing for a computationally efficient implementation. Numerical simulations illustrate how the proposed method offers a performance similar to the group-LASSO for incoherent dictionaries, and superior performance for coherent dictionaries.
|
|
2. |
- Kronvall, Ted, et al.
(författare)
-
Online group-sparse estimation using the covariance fitting criterion
- 2017
-
Ingår i: 25th European Signal Processing Conference, EUSIPCO 2017. - 9780992862671 ; , s. 2101-2105
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we present a time-recursive implementation of a recent hyperparameter-free group-sparse estimation technique. This is achieved by reformulating the original method, termed group-SPICE, as a square-root group-LASSO with a suitable regularization level, for which a time-recursive implementation is derived. Using a proximal gradient step for lowering the computational cost, the proposed method may effectively cope with data sequences consisting of both stationary and non-stationary signals, such as transients, and/or amplitude modulated signals. Numerical examples illustrates the efficacy of the proposed method for both coherent Gaussian dictionaries and for the multi-pitch estimation problem.
|
|
3. |
- Sward, Johan, et al.
(författare)
-
A generalization of the sparse iterative covariance-based estimator
- 2017
-
Ingår i: 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. - 9781509041176 ; , s. 3954-3958
-
Konferensbidrag (refereegranskat)abstract
- In this work, we extend the popular sparse iterative covariance-based estimator (SPICE) by generalizing the formulation to allow for different norm constraint on the signal and noise parameters in the covariance model. For any choice of norms, the resulting generalized SPICE method enjoys the same benefits as the regular SPICE method, including being hyperparameter free, although the choice of norm is shown to govern the sparsity in the resulting solution. Furthermore, we show that there is a connection between the generalized SPICE and a penalized regression problem, both for the case were one allows the noise parameters to differ for each sample, and when treating each noise parameter as being equal. We examine the performance of the method for different choices of norms, and compare the results to the original SPICE method, showing the benefits of using the generalized version. We also provide a way of solving the generalized SPICE using a gridless method, which solves a semi-definite programming problem.
|
|