Abstract
We study sparse approximate solutions to convex optimization problems. It is known that in many engineering applications researchers are interested in an approximate solution of an optimization problem as a linear combination of a few elements from a given system of elements. There is an increasing interest in building such sparse approximate solutions using different greedy-type algorithms. The problem of approximation of a given element of a Banach space by linear combinations of elements from a given system (dictionary) is well studied in nonlinear approximation theory. At first glance, the settings of approximation and optimization problems are very different. In the approximation problem, an element is given and our task is to find a sparse approximation of it. In optimization theory, an energy function is given and we should find an approximate sparse solution to the minimization problem. It turns out that the same technique can be used for solving both problems. We show how the technique developed in nonlinear approximation theory, in particular the greedy approximation technique, can be adjusted for finding a sparse solution of an optimization problem.
Similar content being viewed by others
References
Bahmani, S., Raj, B., Boufounos, P.: Greedy sparsity-constrained optimization. arXiv:1203.5483v3 [stat.ML]. 6 Jan 2013
Barron, A.R.: Universal approximation bounds for superposition of \(n\) sigmoidal functions. IEEE Trans. Inf. Theory 39, 930–945 (1993)
Beck, A., Eldar, Y.C.: Sparsity constrained nonlinear optimization: optimality conditions and algorithms. arXiv:1203.4580v1 [cs.IT]. 20 March 2012
Blumensath, T.: Compressed sensing with nonlinear observations. Preprint 1–9 (2010)
Blumensath, T.: Compressed sensing with nonlinear observations and related nonlinear optimization problems. arXiv:1205.1650v1 [cs.IT]. 8 May 2012
Blumensath, T., Davies, M.E.: Gradient pursuits. IEEE Trans. Signal Process. 56, 2370–2382 (2008)
Blumensath, T., Davies, M.E.: Stagewise weak gradient pursuits. IEEE Trans. Signal Process. 57, 4333–4346 (2009)
Blumensath, T., Davies, M.E.: Iterative hard thresholding for compressed sensing. Appl. Comput. Harmon. Anal. 27, 265–274 (2009)
Borwein, J.M., Lewis, A.S.: Convex Analysis and Nonlinear Optimization: Theory and Examples. Canadian Mathematical Society, Springer, Berlin (2006)
Borwein, J., Guirao, A.J., Hajek, P., Vanderwerff, J.: Uniformly convex functions an Banach spaces. Proc. Am. Math. Soc. 137(3), 1081–1091 (2009)
Chandrasekaran, V., Recht, B., Parrilo, P.A., Willsky, A.S.: The convex geometry of linear inverse problems. In: Proceedings of the 48th Annual Allerton Conference on Communication, Control and Computing, pp. 699–703 (2010)
Clarkson, K.L.: Coresets, sparse greedy approximation, and the Frank-Wolfe algorithm. ACM Trans. Algorithms 6, 63 (2010)
Dai, W., Milenkovic, O.: Subspace pursuit for compressive sensing signal reconstruction. IEEE Trans. Inf. Theory 55, 2230–2249 (2009)
Demyanov, V.F., Rubinov, A.M.: Approximate Methods in Optimization Problems. Elsevier, Amsterdam (1970)
DeVore, R.A.: Nonlinear approximation. Acta Numer. 7, 51–150 (1998)
DeVore, R.A., Temlyakov, V.N.: Convex Optimization on Banach Spaces. arXiv:1401.0334v1 [stat.ML]. 1 Jan 2014
Dudik, M., Harchaoui, Z., Malick, J.: Lifted coordinate descent for learning with trace-norm regularization. In: AISTATS (2012)
Dunn, J.C., Harshbarger, S.: Conditional gradient algorithms with open loop step size rules. J. Math. Anal. Appl. 62, 432–444 (1978)
Elad, M., Matalon, B., Zibulevsky, M.: Coordinate and subspace optimization methods for linear least squares with non-quadratic regularization. Appl. Comput. Harmonic Anal. 23, 346–367 (2007)
Figueiredo, M.A.T., Nowak, R.D., Wright, S.J.: Gradient projection for sparse reconstruction: application to compressed sensing and other inverse problems. IEEE Sel. Top. Signal Process. 1, 586–597 (2007)
Frank, M., Wolfe, P.: An algorithm for quadratic programming. Nav. Res. Logist. Q. 3, 95–110 (1956)
Jaggi, M.: Sparse Convex Optimization Methods for Machine Learning, PhD thesis, ETH Zürich, (2011)
Jaggi, M.: Revisiting Frank-Wolfe: projection-free sparse convex optimization. In: Proceedings of the 30th International Conference on Machine Learning, Atlanta, Georgia (2013)
Jaggi, M., Sulovský, M.: A Simple Algorithm for Nuclear Norm Regularized Problems. ICML (2010)
Jones, L.K.: A simple lemma on greedy approximation in Hilbert space and convergence rates for projection pursuit regression and neural network training. Ann. Statist. 20, 608–613 (1992)
Karmanov, V.G.: Mathematical Programming. Mir Publishers, Moscow (1989)
Kashin, B.S.: Widths of certain finite-dimensional sets and classes of smooth functions. Izv. Acad. Nauk SSSR Ser. Mat 41, 334–351 (1977)
Livshitz, E., Temlyakov, V.: Sparse Approximation and Recovery by Greedy Algorithms. arXiv:1303.3595v1 [math.NA]. 14 March 2013
Needell, D., Tropp, J.A.: CoSaMP: iterative signal recovery from incomplete and inaccurate samples. Appl. Comput. Harmonic Anal. 26, 301–321 (2009)
Needell, D., Vershynin, R.: Uniform uncertainty principle and signal recovery via orthogonal matching pursuit. Found. Comp. Math. 9, 317–334 (2009)
Nesterov, Yu.: Introductory Lectures on Convex Optimization: A Basic Course. Kluwer Academic Publishers, Boston (2004)
Nguyen, H., Petrova, G.: Greedy Strategies for Convex Optimization. arXiv:1401.1754v1 [math.NA]. 8 Jan 2014
Pshenichnyi, B.N., Danilin, YuM: Numerical Methods in Extremal Problems [in Russian]. Nauka, Moscow (1975)
Shalev-Shwartz, S., Srebro, N., Zhang, T.: Trading accuracy for sparsity in optimization problems with sparsity constrains. SIAM J. Optim. 20(6), 2807–2832 (2010)
Temlyakov, V.N.: Weak greedy algorithms. Adv. Comput. Math. 12, 213–227 (2000)
Temlyakov, V.N.: Greedy algorithms in Banach spaces. Adv. Comput. Math. 14, 277–292 (2001)
Temlyakov, V.N.: Greedy-type approximation in Banach spaces and applications. Constr. Approx. 21, 257–292 (2005)
Temlyakov, V.N.: Relaxation in greedy approximation. Constr. Approx. 28, 1–25 (2008)
Temlyakov, V.N.: Greedy approximation. Acta Numer. 17, 235–409 (2008)
Temlyakov, V.N.: Greedy Approximation. Cambridge University Press, Cambridge (2011)
Temlyakov, V.N.: Greedy Approximation in Convex Optimization. arXiv:1206.0392v1 [stat.ML]. 2 June 2012 (see also IMI Preprint, 2012:03, 1–25).
Temlyakov, V.N.: Greedy expansions in convex optimization. In: Proceedings of the Steklov Institute of Mathematics, vol. 284, pp. 244–262, (2014) (see also arXiv:1206.0393v1 [stat.ML]. 2 Jun 2012)
Temlyakov, V.N.: Sparse Approximation and Recovery by Greedy Algorithms in Banach Spaces. arXiv:1303.6811v1 [stat.ML]. 27 March 2013
Temlyakov, V.N.: Chebyshev Greedy Algorithm in Convex Optimization. arXiv:1312.1244v1 [stat.ML]. 4 Dec 2013
Tewari, A., Ravikumar, P., Dhillon, I.S.: Greedy Algorithms for Structurally Constrained High Dimensional Problems, prerint 1–10 (2012)
Zhang, T.: Sequential greedy approximation for certain convex optimization problems. IEEE Trans. Inf. Theory 49(3), 682–691 (2003)
Zhang, T.: Adaptive forward–backward greedy algorithm for learning sparse representations. IEEE Trans. Inf. Theory 57, 4689–4708 (2011)
Zhang, T.: Sparse recovery with orthogonal matching pursuit under RIP. IEEE Trans. Inf. Theory 57, 6215–6221 (2011)
Acknowledgments
This paper was motivated by the IMA Annual Program Workshop “Machine Learning: Theory and Computation” (March 26–30, 2012), in particular, by talks of Steve Wright and Pradeep Ravikumar. The author is very thankful to Arkadi Nemirovski for an interesting discussion of the results and for his remarks. The author is also very grateful to the referees and the editor whose detailed comments helped to substantially improve the presentation of the results.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Joel A. Tropp.
Research was supported by NSF Grant DMS-1160841.
Rights and permissions
About this article
Cite this article
Temlyakov, V.N. Greedy Approximation in Convex Optimization. Constr Approx 41, 269–296 (2015). https://doi.org/10.1007/s00365-014-9272-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00365-014-9272-0