Abstract
Recently, the optimization methods for computing higher-order critical points of nonconvex problems attract growing research interest (Anandkumar Conference on Learning Theory 81-102, 2016), (Cartis Found Comput Math 18:1073-1107, 2018), (Cartis SIAM J Optim 30:513-541, 2020), (Chen Math Program 187:47-78, 2021) , as they are able to exclude the so-called degenerate saddle points and reach a solution with better quality. Despite theoretical developments in (Anandkumar Conference on Learning Theory 81-102, 2016), (Cartis Found Comput Math 18:1073-1107, 2018), (Cartis SIAM J Optim 30:513-541, 2020), (Chen Math Program 187:47-78, 2021) , the corresponding numerical experiments are missing. This paper proposes an implementable higher-order method, named adaptive high order method (AHOM), to find the third-order critical points. AHOM is achieved by solving an “easier” subproblem and incorporating the adaptive strategy of parameter-tuning in each iteration of the algorithm. The iteration complexity of the proposed method is established. Some preliminary numerical results are provided to show that AHOM can escape from the degenerate saddle points, where the second-order method could possibly get stuck.
Similar content being viewed by others
References
Amaral, V., Andreani, R., Birgin, E., Marcondes, D., Martínez, J.: On complexity and convergence of high-order coordinate descent algorithms for smooth nonconvex box-constrained minimization. (2020) arXiv preprint arXiv:2009.01811
Anandkumar, A., Ge, R.: Efficient approaches for escaping higher order saddle points in non-convex optimization. In: Conference on Learning Theory, pp 81–102 (2016)
Baes, M.: Estimate sequence methods: extensions and approximations. Institute for Operations Research, ETH, Z\(^{..1}\)rich, Switzerland (2009)
Bellavia, S., Gurioli, G., Morini, B., Toint, P.L.: Adaptive regularization algorithms with inexact evaluations for nonconvex optimization. SIAM J. Optim. 29(4), 2881–2915 (2019)
Birgin, E., Krejić, N., Martínez, J.: Economic inexact restoration for derivative-free expensive function minimization and applications. (2020) arXiv preprint arXiv:2009.09062
Birgin, E.G., Gardenghi, J., Martínez, J.M., Santos, S.A.: On the use of third-order models with fourth-order regularization for unconstrained optimization. Optim. Lett. 14, 815–838 (2020)
Birgin, E.G., Gardenghi, J., Martinez, J.M., Santos, S.A., Toint, P.L.: Evaluation complexity for nonlinear constrained optimization using unscaled KKT conditions and high-order models. SIAM J. Optim. 26(2), 951–967 (2016)
Birgin, E.G., Gardenghi, J., Martínez, J.M., Santos, S.A., Toint, P.L.: Worst-case evaluation complexity for unconstrained nonlinear optimization using high-order regularized models. Math. Prog. 163(1–2), 359–368 (2017)
Bubeck, S., Jiang, Q., Lee, Y.T., Li, Y., Sidford, A.: Near-optimal method for highly smooth convex optimization. In: Conference on Learning Theory, pp. 492–507. PMLR (2019)
Carmon, Y., Duchi, J.C., Hinder, O., Sidford, A.: Lower bounds for finding stationary points I. Math. Prog. 184, 1–50 (2017)
Carmon, Y., Duchi, J.C., Hinder, O., Sidford, A.: Accelerated methods for nonconvex optimization. SIAM J. Optim. 28(2), 1751–1772 (2018)
Cartis, C., Gould, N.I., Toint, P.L.: Adaptive cubic regularisation methods for unconstrained optimization. part I: motivation, convergence and numerical results. Math. Program. 127(2), 245–295 (2011)
Cartis, C., Gould, N.I., Toint, P.L.: Adaptive cubic regularisation methods for unconstrained optimization. part II: worst-case function-and derivative-evaluation complexity. Math. Program. 130(2), 295–319 (2011)
Cartis, C., Gould, N.I., Toint, P.L.: Second-order optimality and beyond: characterization and evaluation complexity in convexly constrained nonlinear optimization. Found. Comput. Math. 18(5), 1073–1107 (2018)
Cartis, C., Gould, N.I., Toint, P.L.: A concise second-order complexity analysis for unconstrained optimization using high-order regularized models. Opt. Methods Softw. 35(2), 243–256 (2020)
Cartis, C., Gould, N.I., Toint, P.L.: Sharp worst-case evaluation complexity bounds for arbitrary-order nonconvex optimization with inexpensive constraints. SIAM J. Optim. 30(1), 513–541 (2020)
Chen, X., Jiang, B., Lin, T., Zhang, S.: Accelerating adaptive cubic regularization of Newton’s method via random sampling. J. Mach. Learn. Res. (2022)
Chen, X., Toint, P.L.: High-order evaluation complexity for convexly-constrained optimization with non-lipschitzian group sparsity terms. Math. Program. 187(1), 47–78 (2021)
Chen, X., Toint, P.L., Wang, H.: Complexity of partially separable convexly constrained optimization with Non-Lipschitzian singularities. SIAM J. Optim. 29(1), 874–903 (2019)
Curtis, F.E., Robinson, D.P., Samadi, M.: An inexact regularized Newton framework with a worst-case iteration complexity of for nonconvex optimization. IMA J. Numer. Anal. 39(3), 1296–1327 (2018)
Cushing, J.M.: Extremal tests for scalar functions of several real variables at degenerate critical points. Aequationes Math. 13(1–2), 89–96 (1975)
Duchi, J., Hazan, E., Singer, Y.: Adaptive subgradient methods for online learning and stochastic optimization. J. Mach. Learn. Res. 12, 2121–2159 (2011)
Gasnikov, A., Kovalev, D., Mohhamed, A., Chernousova, E.: The global rate of convergence for optimal tensor methods in smooth convex optimization. (2018) arXiv preprint arXiv:1809.00382
Ghadimi, S., Lan, G., Zhang, H.: Generalized uniformly optimal methods for nonlinear programming. J. Sci. Comput. 79(3), 1854–1881 (2019)
Gould, N.I., Lucidi, S., Roma, M., Toint, P.L.: Solving the trust-region subproblem using the lanczos method. SIAM J. Optim. 9(2), 504–525 (1999)
Gould, N.I., Rees, T., Scott, J.A.: Convergence and evaluation-complexity analysis of a regularized tensor-Newton method for solving nonlinear least-squares problems. Comput. Optim. Appl. 73(1), 1–35 (2019)
Grapiglia, G.N., Nesterov, Y.: Tensor methods for finding approximate stationary points of convex functions. Optim. Methods Softw. pp. 1–34 (2020)
Grapiglia, G.N., Nesterov, Y.: Tensor methods for minimizing convex functions with hölder continuous higher-order derivatives. SIAM J. Optim. 30(4), 2750–2779 (2020)
Gratton, S., Simon, E., Toint, P.L.: An algorithm for the minimization of nonsmooth nonconvex functions using inexact evaluations and its worst-case complexity. Math. Program. 187(1), 1–24 (2021)
Jiang, B., Lin, T., Ma, S., Zhang, S.: Structured nonconvex and nonsmooth optimization: algorithms and iteration complexity analysis. Comput. Optim. Appl. 72, 115–157 (2019)
Jiang, B., Lin, T., Zhang, S.: A unified adaptive tensor approximation scheme to accelerate composite convex optimization. SIAM J. Optim. 30, 2897–2926 (2020)
Jiang, B., Wang, H., Zhang, S.: An optimal high-order tensor method for convex optimization. Math. Oper. Res. 46, 1390–1412 (2021)
Lucchi, A., Kohler, J.: A stochastic tensor method for non-convex optimization. (2019) arXiv preprint arXiv:1911.10367
Martínez, J.M.: On high-order model regularization for constrained optimization. SIAM J. Optim. 27(4), 2447–2458 (2017)
Mason, L., Baxter, J., Bartlett, P., Frean, M.: Boosting algorithms as gradient descent. In: Solla S., Leen T., Müller K. (eds.) Advances in neural information processing systems, vol 12, pp. 512–518 (1999)
Motzkin, T.S.: The arithmetic-geometric inequality. Inequalities (Proc. Sympos. Wright-Patterson Air Force Base, Ohio, 1965) pp. 205–224 (1967)
Murty, K., Kabadi, S.: Some np-complete problems in quadratic and nonlinear programming. Math. Program. 39, 117–129 (1987)
Nesterov, Y.: Introductory lectures on convex optimization a basic course. Appl. Optim. 87(5), 236 (2004)
Nesterov, Y.: Accelerating the cubic regularization of Newton’s method on convex problems. Math. Program. 112(1), 159–181 (2008)
Nesterov, Y.: Implementable tensor methods in unconstrained convex optimization. Math. Program. 186(1), 157–183 (2021)
Nesterov, Y., Polyak, B.T.: Cubic regularization of Newton method and its global performance. Math. Program. 108(1), 177–205 (2006)
Nie, J.: The hierarchy of local minimums in polynomial optimization. Math. Program. 151(2), 555–583 (2015)
Udell, M., Boyd, S.: Maximizing a sum of sigmoids. Optim. Eng.pp. 1–25 (2013)
Zheng, Y., Zheng, B.: A modified adaptive cubic regularization method for large-scale unconstrained optimization problem. (2019) arXiv preprint arXiv:1904.07440
Acknowledgements
We would like to thank the two anonymous referees for their insightful comments, and we would like to also thank Professor Qi Deng at Shanghai University of Finance and Economics for the discussion on the numerical experiment of this paper.
Author information
Authors and Affiliations
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Research supported by NSFC Grants 72171141, NSFC Grants 71971132, NSFC Grants 72150001, NSFC Grants 11831002, GIFSUFE Grants CXJJ-2019-391, and Program for Innovative Research Team of Shanghai University of Finance and Economics.
Rights and permissions
About this article
Cite this article
Zhu, X., Han, J. & Jiang, B. An adaptive high order method for finding third-order critical points of nonconvex optimization. J Glob Optim 84, 369–392 (2022). https://doi.org/10.1007/s10898-022-01151-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10898-022-01151-1
Keywords
- Continuous optimization
- Nonconvex optimization
- Adaptive algorithm
- Higher order method
- Third-order critical points