Regularized Decomposition of High-Dimensional Multistage Stochastic Programs with Markov Uncertainty

We develop a quadratic regularization approach for the solution of high-dimensional multistage stochastic optimization problems characterized by a potentially large number of time periods/stages (e.g. hundreds), a high-dimensional resource state variable, and a Markov information process. The resulting algorithms are shown to converge to an optimal policy after a finite number of iterations under mild technical assumptions. Computational experiments are conducted using the setting of optimizing energy storage over a large transmission grid, which motivates both the spatial and temporal dimensions of our problem. Our numerical results indicate that the proposed methods exhibit significantly faster convergence than their classical counterparts, with greater gains observed for higher-dimensional problems.

[1]  A. Gjelsvik,et al.  Long- and Medium-term Operations Planning and Stochastic Modelling in Hydro-dominated Power Systems Based on Stochastic Dual Dynamic Programming , 2010 .

[2]  Panos M. Pardalos,et al.  Approximate dynamic programming: solving the curses of dimensionality , 2009, Optim. Methods Softw..

[3]  Andrzej Ruszczynski,et al.  A regularized decomposition method for minimizing a sum of polyhedral functions , 1986, Math. Program..

[4]  Yurii Nesterov,et al.  Introductory Lectures on Convex Optimization - A Basic Course , 2014, Applied Optimization.

[5]  Andy Philpott,et al.  Solving long-term hydro-thermal scheduling problems , 2011 .

[6]  Alexander Shapiro,et al.  Lectures on Stochastic Programming: Modeling and Theory , 2009 .

[7]  Yurii Nesterov,et al.  Interior-point polynomial algorithms in convex programming , 1994, Siam studies in applied mathematics.

[8]  Vitor L. de Matos,et al.  Dynamic sampling algorithms for multi-stage stochastic programs with risk aversion , 2012, Eur. J. Oper. Res..

[9]  M. V. F. Pereira,et al.  Multi-stage stochastic optimization applied to energy planning , 1991, Math. Program..

[10]  John R. Birge,et al.  The Abridged Nested Decomposition Method for Multistage Stochastic Linear Programs with Relatively Complete Recourse , 2006, Algorithmic Oper. Res..

[11]  Andrzej Ruszczyński,et al.  Regularized Decomposition of Stochastic Programs: Algorithmic Techniques and Numerical Results , 1993 .

[12]  Martin L. Puterman,et al.  Markov Decision Processes: Discrete Stochastic Dynamic Programming , 1994 .

[13]  J. M. Damázio,et al.  The use of PAR(p) model in the stochastic dual dynamic programming optimization scheme used in the operation planning of the Brazilian hydropower system , 2005, 2004 International Conference on Probabilistic Methods Applied to Power Systems.

[14]  R. Rockafellar Monotone Operators and the Proximal Point Algorithm , 1976 .

[15]  Monia Rekik,et al.  Stochastic dual dynamic programming for transportation planning under demand uncertainty , 2013, 2013 International Conference on Advanced Logistics and Transport.

[16]  Birger Mo,et al.  Integrated risk management of hydro power scheduling and contract management , 2001 .

[17]  D. Morton,et al.  Assessing policy quality in multi-stage stochastic programming , 2004 .

[18]  Warren B. Powell,et al.  Adaptive stepsizes for recursive estimation with applications in approximate dynamic programming , 2006, Machine Learning.

[19]  Hande Y. Benson Interior‐Point Linear Programming Solvers , 2011 .

[20]  Andrzej Ruszczynski,et al.  Two-Stage Portfolio Optimization with Higher-Order Conditional Measures of Risk , 2012 .

[21]  Tito Homem-de-Mello,et al.  Sampling strategies and stopping criteria for stochastic dual dynamic programming: a case study in long-term hydrothermal scheduling , 2011 .

[22]  Andrzej Ruszczynski,et al.  Time-consistent approximations of risk-averse multistage stochastic optimization problems , 2015, Math. Program..

[23]  Zhihong Zhou,et al.  Multistage Stochastic Decomposition: A Bridge between Stochastic Programming and Approximate Dynamic Programming , 2014, SIAM J. Optim..

[24]  Andrzej Ruszczynski,et al.  Decomposition methods in stochastic programming , 1997, Math. Program..

[25]  Ying Xiong Nonlinear Optimization , 2014 .

[26]  Jacques F. Benders,et al.  Partitioning procedures for solving mixed-variables programming problems , 2005, Comput. Manag. Sci..

[27]  Warren B. Powell,et al.  What you should know about approximate dynamic programming , 2009, Naval Research Logistics (NRL).

[28]  Václav Kozmı́k Multistage risk-averse asset allocation with transaction costs , 2012 .

[29]  I. Lustig,et al.  Computational experience with a primal-dual interior point method for linear programming , 1991 .

[30]  Erlon Cristian Finardi,et al.  Improving the performance of Stochastic Dual Dynamic Programming , 2015, J. Comput. Appl. Math..

[31]  Julia L. Higle,et al.  Finite master programs in regularized stochastic decomposition , 1994, Math. Program..

[32]  R. Wets,et al.  L-SHAPED LINEAR PROGRAMS WITH APPLICATIONS TO OPTIMAL CONTROL AND STOCHASTIC PROGRAMMING. , 1969 .

[33]  V. Kozmík,et al.  Risk-Averse Stochastic Dual Dynamic Programming , 2013 .

[34]  Shirley Dex,et al.  JR 旅客販売総合システム(マルス)における運用及び管理について , 1991 .

[35]  Alexander Shapiro,et al.  Analysis of stochastic dual dynamic programming method , 2011, Eur. J. Oper. Res..

[36]  Stefan Minner,et al.  Optimizing Trading Decisions for Hydro Storage Systems Using Approximate Dual Dynamic Programming , 2013, Oper. Res..

[37]  Alexander Shapiro,et al.  Risk neutral and risk averse Stochastic Dual Dynamic Programming method , 2013, Eur. J. Oper. Res..

[38]  Yada Zhu,et al.  Multiobjective Optimization of a Port-of-Entry Inspection Policy , 2010, IEEE Transactions on Automation Science and Engineering.

[39]  Robert J. Vanderbei,et al.  An Interior-Point Method for Semidefinite Programming , 1996, SIAM J. Optim..

[40]  Sanjay Mehrotra,et al.  On the Implementation of a Primal-Dual Interior Point Method , 1992, SIAM J. Optim..

[41]  Andrew B. Philpott,et al.  On the convergence of stochastic dual dynamic programming and related methods , 2008, Oper. Res. Lett..

[42]  Pierre Girardeau,et al.  On the Convergence of Decomposition Methods for Multistage Stochastic Convex Programs , 2015, Math. Oper. Res..

[43]  Nikki Newham,et al.  Power System Investment Planning using Stochastic Dual Dynamic Programming , 2008 .

[44]  Andy Philpott,et al.  On the Convergence of Sampling-Based Decomposition Algorithms for Multistage Stochastic Programs , 2005 .

[45]  John R. Birge,et al.  Decomposition and Partitioning Methods for Multistage Stochastic Linear Programs , 1985, Oper. Res..

[46]  A. Ruszczynski,et al.  Accelerating the regularized decomposition method for two stage stochastic linear problems , 1997 .

[47]  Stephen P. Boyd,et al.  An Interior-Point Method for Large-Scale $\ell_1$-Regularized Least Squares , 2007, IEEE Journal of Selected Topics in Signal Processing.

[48]  J. E. Kelley,et al.  The Cutting-Plane Method for Solving Convex Programs , 1960 .

[49]  Laurent Pfeiffer,et al.  Two methods of pruning Benders' cuts and their application to the management of a gas portfolio , 2012 .

[50]  Andrzej Ruszczynski,et al.  Risk-averse dynamic programming for Markov decision processes , 2010, Math. Program..

[51]  Warrren B Powell,et al.  Convergent Cutting-Plane and Partial-Sampling Algorithm for Multistage Stochastic Linear Programs with Recourse , 1999 .

[52]  Andrzej Ruszczynski,et al.  Two-stage portfolio optimization with higher-order conditional measures of risk , 2012, Ann. Oper. Res..

[53]  Gerd Infanger,et al.  Cut sharing for multistage stochastic linear programs with interstage dependency , 1996, Math. Program..

[54]  Anderson Rodrigo de Queiroz,et al.  Sharing cuts under aggregated forecasts when decomposing multi-stage stochastic programs , 2013, Oper. Res. Lett..

[55]  Stephen J. Wright Primal-Dual Interior-Point Methods , 1997, Other Titles in Applied Mathematics.

[56]  David P. Morton,et al.  An enhanced decomposition algorithm for multistage stochastic hydroelectric scheduling , 1996, Ann. Oper. Res..