Abstract
In Mitsos et al. (SIAM Journal on Optimization 20(2):573–601, 2009), a method similar to Algorithmic Differentiation (AD) is presented which allows the propagation of, in general nondifferentiable, McCormick relaxations (McCormick, Mathematical Programming 10(2):147–175, 1976; Steihaug, Twelfth Euro AD Workshop, Berlin, 2011) of factorable functions and of the corresponding subgradients in tangent-linear mode. Subgradients are natural extensions of “usual” derivatives which allow the application of derivative-based methods to possibly nondifferentiable convex and concave functions. The software package libMC (Mitsos et al. SIAM Journal on Optimization 20(2):573–601, 2009) performs the automatic propagation of the relaxation and of corresponding subgradients based on the principles of tangent-linear mode AD by overloading. Similar ideas have been ported to Fortran yielding modMC as part of our ongoing collaboration with the authors of Mitsos et al. (SIAM Journal on Optimization 20(2):573–601, 2009). In this article an adjoint method for the computation of subgradients for McCormick relaxations is presented. A corresponding implementation by overloading in Fortran is provided in the form of amodMC. The calculated subgradients are used in a deterministic global optimization algorithm based on a branch-and-bound method. The superiority of adjoint over tangent-linear mode is illustrated by two examples.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
References
Bauer, F.: Computational graphs and rounding error. SIAM Journal on Numerical Analysis 11(1), 87–96 (1974). DOI 10.1137/0711010
Corbett, C., Naumann, U.: Demonstration of a branch-and-bound algorithm for global optimization using McCormick relaxations. Tech. Rep. AIB-2011-24, RWTH Aachen (2011). URL http://aib.informatik.rwth-aachen.de/2011/2011-24.pdf
Griewank, A.: Piecewise linearization via Algorithmic Differentiation. Twelfth Euro AD Workshop, Berlin (2011)
Griewank, A., Walther, A.: Evaluating Derivatives: Principles and Techniques of Algorithmic Differentiation, 2nd edn. No. 105 in Other Titles in Applied Mathematics. SIAM, Philadelphia, PA (2008). URL http://www.ec-securehost.com/SIAM/OT105.html
Hiriart-Urruty, J., Lemaréchal, C.: Fundamentals of Convex Analysis. Springer-Verlag (2001)
McCormick, G.P.: Computability of global solutions to factorable nonconvex programs: Part I. Convex underestimating problems. Mathematical Programming 10(2), 147–175 (1976)
Mitsos, A., Chachuat, B., Barton, P.I.: McCormick-based relaxations of algorithms. SIAM Journal on Optimization 20(2), 573–601 (2009)
Naumann, U.: The Art of Differentiating Computer Programs. An Introduction to Algorithmic Differentiation. Software, Environments, and Tools. SIAM (2011)
Schnurr, M.: The automatic computation of second-order slope tuples for some nonsmooth functions (2007)
Steihaug, T.: Factorable programming revisited. Twelfth Euro AD Workshop, Berlin (2011)
Acknowledgements
Markus Beckers is supported by the German Research School for Simulation Sciences. Viktor Mosenkis is supported by the German Science Foundation (DFG grant No. 487/2-1).
We would like to acknowledge several fruitful discussions on the subject with Alexander Mitsos from MIT’s Mechanical Engineering.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Beckers, M., Mosenkis, V., Naumann, U. (2012). Adjoint Mode Computation of Subgradients for McCormick Relaxations. In: Forth, S., Hovland, P., Phipps, E., Utke, J., Walther, A. (eds) Recent Advances in Algorithmic Differentiation. Lecture Notes in Computational Science and Engineering, vol 87. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-30023-3_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-30023-3_10
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-30022-6
Online ISBN: 978-3-642-30023-3
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)