{"type":"ScholarlyArticle","title":"Efficient Inexact Proximal Gradient Algorithm for Nonconvex Problems","authors":[{"name":"Quanming Yao","slug":"quanming-yao","given":"Quanming","family":"Yao","affiliation":"Microsoft Research , Beijing , China"},{"name":"James T. Kwok","slug":"james-t-kwok","given":"James T.","family":"Kwok","affiliation":"Microsoft Research , Beijing , China"},{"name":"Fei Gao","slug":"fei-gao","given":"Fei","family":"Gao","affiliation":"Hong Kong University of Science and Technology , Clear Water Bay , Hong Kong"},{"name":"Wei Chen","slug":"wei-chen","given":"Wei","family":"Chen","affiliation":"Hong Kong University of Science and Technology , Clear Water Bay , Hong Kong"},{"name":"Tie-Yan Liu","slug":"tie-yan-liu","given":"Tie-Yan","family":"Liu","affiliation":""}],"abstract":"While proximal gradient algorithm is originally designed for convex optimization, several variants have been recently proposed for nonconvex problems. Among them, nmAPG  [Li and Lin, 2015] is the state-of-art. However, it is inefficient when the proximal step does not have closed-form solution, or such solution exists but is expensive, as it requires more than one proximal steps to be exactly solved in each iteration.  In this paper, we propose an efficient accelerate proximal gradient (niAPG) algorithm for nonconvex problems. In each iteration, it requires only one inexact (less expensive) proximal step. Convergence to a critical point is still guaranteed, and a O(1/k) convergence rate is derived. Experiments on image inpainting and matrix completion problems demonstrate that the proposed algorithm has comparable performance as the state-of-the-art, but is much faster.","doi":"10.24963/ijcai.2017/462","url":"https://ijcai.org/proceedings/2017/462","pdf":"https://ijcai.org/proceedings/2017/0462.pdf","markdown":"https://ijcai.org/proceedings/2017/462.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/462","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"Machine Learning S-Z","track":"Main track","number":462,"pages":{"first":3308,"last":3314},"keywords":["Machine Learning: Structured Learning","Machine Learning: Machine Learning"],"videos":[],"references":[{"position":1,"raw":"[Attouch et al., 2013] H. Attouch, J. Bolte, and B.F. Svaiter. Convergence of descent methods for semi-algebraic and tame problems: Proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods. Mathe- matical Programming, 137(1-2):91-129, 2013.","title":"Convergence of descent methods for semi-algebraic and tame problems: Proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods","authors":"H Attouch, J Bolte, B F Svaiter","year":"2013","venue":"Mathematical Programming","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Beck and Teboulle, 2009a] A. Beck and M. Teboulle. Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems. IEEE Transac- tions on Image Processing, 18(11):2419-2434, 2009.","title":"Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems","authors":"Teboulle Beck, ] A Beck, M Teboulle","year":"2009","venue":"IEEE Transactions on Image Processing","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Beck and Teboulle, 2009b] A. Beck and M. Teboulle. A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM Journal on Imaging Sciences, 2(1):183-202, 2009.","title":"A fast iterative shrinkage-thresholding algorithm for linear inverse problems","authors":"Teboulle Beck, A Beck, M Teboulle","year":"2009","venue":"SIAM Journal on Imaging Sciences","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Boyd et al., 2011] S. Boyd, N. Parikh, E. Chu, B. Peleato, and J. Eckstein. Distributed optimization and statistical learning via the alternating direction method of multiplier- s. Foundations and Trends in Machine Learning, 3(1):1- 122, 2011.","title":"Distributed optimization and statistical learning via the alternating direction method of multipliers","authors":"S Boyd, N Parikh, E Chu, B Peleato, J Eckstein","year":"2011","venue":"Foundations and Trends in Machine Learning","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Candès et al., 2008] E.J. Candès, M.B. Wakin, and S. Boyd. Enhancing sparsity by reweighted ℓ 1 minimization. Jour- nal of Fourier Analysis and Applications, 14(5-6):877- 905, 2008.","title":"Enhancing sparsity by reweighted ℓ 1 minimization","authors":"E J Candès, M B Wakin, S Boyd","year":"2008","venue":"Journal of Fourier Analysis and Applications","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Ghadimi and Lan, 2015] S. Ghadimi and G. Lan. Accel- erated gradient methods for nonconvex nonlinear and stochastic programming. Mathematical Programming, 156(1-2):1-41, 2015.","title":"Accelerated gradient methods for nonconvex nonlinear and stochastic programming","authors":"S Ghadimi, G Lan","year":"2015","venue":"Mathematical Programming","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Gong et al., 2013] P. Gong, C. Zhang, Z. Lu, J. Huang, and J. Ye. A general iterative shrinkage and thresholding algo- rithm for non-convex regularized optimization problems. In ICML, pages 37-45, 2013.","title":"A general iterative shrinkage and thresholding algorithm for non-convex regularized optimization problems","authors":"P Gong, C Zhang, Z Lu, J Huang, J Ye","year":"2013","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Grippo and Sciandrone, 2002] L. Grippo and M. Scian- drone. Nonmonotone globalization techniques for the Barzilai-Borwein gradient method. Computational Opti- mization and Applications, 23(2):143-169, 2002.","title":"Nonmonotone globalization techniques for the Barzilai-Borwein gradient method","authors":"L Grippo, M Sciandrone","year":"2002","venue":"Computational Optimization and Applications","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Halko et al., 2011] N. Halko, P.G. Martinsson, and J.A. Tropp. Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompo- sitions. SIAM Review, 53(2):217-288, 2011.","title":"Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions","authors":"N Halko, P G Martinsson, J A Tropp","year":"2011","venue":"SIAM Review","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Hastie et al., 2015] T. Hastie, R. Mazumder, J.D. Lee, and R. Zadeh. Matrix completion and low-rank SVD via fast alternating least squares. Journal of Machine Learning Research, 16:3367-3402, 2015.","title":"Matrix completion and low-rank SVD via fast alternating least squares","authors":"T Hastie, R Mazumder, J D Lee, R Zadeh","year":"2015","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Hsieh and Olsen, 2014] C.J. Hsieh and P. Olsen. Nuclear norm minimization via active subspace selection. In ICML, pages 575-583, 2014.","title":"Nuclear norm minimization via active subspace selection","authors":"C J Hsieh, P Olsen","year":"2014","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Larsen, 1998] R.M. Larsen. Lanczos bidiagonalization with partial reorthogonalization. DAIMI PB-357, Department of Computer Science, Aarhus University, 1998.","title":"Lanczos bidiagonalization with partial reorthogonalization","authors":"R M Larsen","year":"1998","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Li and Lin, 2015] H. Li and Z. Lin. Accelerated proximal gradient methods for nonconvex programming. In NIPS, pages 379-387, 2015.","title":"Accelerated proximal gradient methods for nonconvex programming","authors":"H Li, Z Lin","year":"2015","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Lu et al., 2014] C. Lu, J. Tang, S Yan, and Z Lin. Gener- alized nonconvex nonsmooth low-rank minimization. In CVPR, pages 4130-4137, 2014.","title":"Generalized nonconvex nonsmooth low-rank minimization","authors":"C Lu, J Tang, Yan, Lin","year":"2014","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Nesterov, 2004] Y. Nesterov. Introductory Lectures on Convex Optimization: A Basic Course. Springer, 2004.","title":"Introductory Lectures on Convex Optimization: A Basic Course","authors":"Y Nesterov","year":"2004","venue":null,"doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Nesterov, 2013] Y. Nesterov. Gradient methods for mini- mizing composite functions. Mathematical Programming, 140(1):125-161, 2013.","title":"Gradient methods for minimizing composite functions","authors":"Y Nesterov","year":"2013","venue":"Mathematical Programming","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Ochs et al., 2014] P. Ochs, Y. Chen, T. Brox, and T. Pock. iPiano: Inertial proximal algorithm for nonconvex opti- mization. SIAM Journal on Imaging Sciences, 7(2):1388- 1419, 2014.","title":"iPiano: Inertial proximal algorithm for nonconvex optimization","authors":"P Ochs, Y Chen, T Brox, T Pock","year":"2014","venue":"SIAM Journal on Imaging Sciences","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Parikh and Boyd, 2014] N. Parikh and S. Boyd. Proximal algorithms. Foundations and Trends in Optimization, 1(3):127-239, 2014.","title":"Proximal algorithms","authors":"N Parikh, S Boyd","year":"2014","venue":"Foundations and Trends in Optimization","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Schmidt et al., 2011] M. Schmidt, N.L. Roux, and F. Bach. Convergence rates of inexact proximal-gradient methods for convex optimization. In NIPS, pages 1458-1466, 2011.","title":"Convergence rates of inexact proximal-gradient methods for convex optimization","authors":"M Schmidt, N L Roux, F Bach","year":"2011","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Wang et al., 2015] Z. Wang, M.J. Lai, Z. Lu, W. Fan, H. Davulcu, and J. Ye. Orthogonal rank-one matrix pursuit for low rank matrix completion. SIAM Journal on Scientific Computing, 37(1), 2015.","title":"Orthogonal rank-one matrix pursuit for low rank matrix completion","authors":"Z Wang, M J Lai, Z Lu, W Fan, H Davulcu, J Ye","year":"2015","venue":"SIAM Journal on Scientific Computing","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Wright et al., 2009] S.J. Wright, R.D. Nowak, and M.A. Figueiredo. Sparse reconstruction by separable ap- proximation. IEEE Transactions on Signal Processing, 57(7):2479-2493, 2009.","title":"Sparse reconstruction by separable approximation","authors":"S J Wright, R D Nowak, M A Figueiredo","year":"2009","venue":"IEEE Transactions on Signal Processing","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Yao and Kwok, 2015] Q. Yao and J.T. Kwok. Accelerated inexact Soft-Impute for fast large-scale matrix completion. In The 24th International Conference on Artificial Intelli- gence, pages 4002-4008, 2015.","title":"Accelerated inexact Soft-Impute for fast large-scale matrix completion","authors":"Kwok Yao, ] Q Yao, J T Kwok","year":"2015","venue":"The 24th International Conference on Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/2015/562"},{"position":23,"raw":"[Yao and Kwok, 2016] Q. Yao and J.T. Kwok. Efficient learning with a family of nonconvex regularizers by redistributing nonconvexity. In ICML, pages 2645-2654, 2016.","title":"Efficient learning with a family of nonconvex regularizers by redistributing nonconvexity","authors":"Kwok Yao, ] Q Yao, J T Kwok","year":"2016","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Yao et al., 2015] Q. Yao, J.T. Kwok, and W. Zhong. Fast low-rank matrix learning with nonconvex regularization. In ICDM, pages 539-548, 2015.","title":"Fast low-rank matrix learning with nonconvex regularization","authors":"Q Yao, J T Kwok, W Zhong","year":"2015","venue":"ICDM","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Yuille and Rangarajan, 2002] A.L. Yuille and A. Rangara- jan. The concave-convex procedure (CCCP). In NIPS, pages 1033-1040, 2002.","title":"The concave-convex procedure (CCCP)","authors":"A L Yuille, A Rangarajan","year":"2002","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhang et al., 2012] X. Zhang, D. Schuurmans, and Y.-L. Yu. Accelerated training for matrix-norm regularization: a boosting approach. In NIPS, pages 2906-2914, 2012.","title":"Accelerated training for matrix-norm regularization: a boosting approach","authors":"Zhang","year":"2012","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhang, 2010] T. Zhang. Analysis of multi-stage convex relaxation for sparse regularization. Journal of Machine Learning Research, 11:1081-1107, 2010.","title":"Analysis of multi-stage convex relaxation for sparse regularization","authors":"T Zhang","year":"2010","venue":"Journal of Machine Learning Research","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Zhao et al., 2015] T. Zhao, Z. Wang, and H. Liu. A nonconvex optimization framework for low rank matrix estimation. In NIPS, pages 559-567, 2015.","title":"A nonconvex optimization framework for low rank matrix estimation","authors":"T Zhao, Z Wang, H Liu","year":"2015","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhong and Kwok, 2014] L.W. Zhong and J.T. Kwok. Gra- dient descent with proximal average for nonconvex and composite regularization. In AAAI, pages 2206-2212, 2014.","title":"Gradient descent with proximal average for nonconvex and composite regularization","authors":"Zhong, ; ] L W Kwok, J T Zhong, Kwok","year":"2014","venue":"AAAI","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}