Adaptive Restart for Accelerated Gradient Schemes
In this paper we introduce a simple heuristic adaptive restart technique that can dramatically improve the convergence rate of accelerated gradient schemes. The analysis of the technique relies on the observation that these schemes exhibit two modes of behavior depending on how much momentum is applied at each iteration. In what we refer to as the ‘high momentum’ regime the iterates generated by an accelerated gradient scheme exhibit a periodic behavior, where the period is proportional to the square root of the local condition number of the objective function. Separately, it is known that the optimal restart interval is proportional to this same quantity. This suggests a restart technique whereby we reset the momentum whenever we observe periodic behavior. We provide a heuristic analysis that suggests that in many cases adaptively restarting allows us to recover the optimal rate of convergence with no prior knowledge of function parameters.
KeywordsConvex optimization First order methods Accelerated gradient schemes
Mathematics Subject Classification80M50 90C06 90C25
- 8.A. Chiang, Fundamental Methods of Mathematical Economics (McGraw-Hill, New York, 1984). Google Scholar
- 11.M. Gu, L. Lim, C. Wu, PARNES: A rapidly convergent algorithm for accurate recovery of sparse and approximately sparse signals. Technical report (2009). arXiv:0911.0492.
- 13.G. Lan, R. Monteiro, Iteration complexity of first-order penalty methods for convex programming. Manuscript, School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, June 2008 Google Scholar
- 14.G. Lan, Z. Lu, R. Monteiro, Primal-dual first-order methods with o(1/ϵ) iteration-complexity for cone programming, Math. Program. 1–29 (2009). Google Scholar
- 15.J. Liu, L. Yuan, J. Ye, An efficient algorithm for a class of fused lasso problems, in Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, July (2010), pp. 323–332. Google Scholar
- 16.A. Nemirovski, Efficient methods in convex programming. Lecture notes (1994). http://www2.isye.gatech.edu/~nemirovs/Lect_EMCO.pdf.
- 17.A. Nemirovski, D. Yudin, Problem Complexity and Method Efficiency in Optimization. Wiley-Interscience Series in Discrete Mathematics (Wiley, New York, 1983). Google Scholar
- 20.Y. Nesterov, Gradient methods for minimizing composite objective function. CORE discussion paper (2007). http://www.ecore.be/DPs/dp_1191313936.pdf.
- 21.J. Nocedal, S. Wright, Numerical Optimization. Springer Series in Operations Research (Springer, Berlin, 2000). Google Scholar
- 22.B. Polyak, Introduction to Optimization. Translations Series in Mathematics and Engineering (Optimization Software, Publications Division, New York, 1987). Google Scholar
- 24.P. Tseng, On accelerated proximal gradient methods for convex-concave optimization (2008). http://pages.cs.wisc.edu/~brecht/cs726docs/Tseng.APG.pdf.