© 2003 Warren B. Powell Slide 1 Approximate Dynamic Programming for High Dimensional Resource Allocation NSF Electric Power workshop November 3, 2003 Warren.

Slides:



Advertisements
Similar presentations
R&D Portfolio Optimization One Stage R&D Portfolio Optimization with an Application to Solid Oxide Fuel Cells Lauren Hannah 1, Warren Powell 1, Jeffrey.
Advertisements

Lect.3 Modeling in The Time Domain Basil Hamed
Monte Carlo Methods and Statistical Physics
Biointelligence Laboratory, Seoul National University
Computer vision: models, learning and inference Chapter 8 Regression.
1 Approximated tracking of multiple non-rigid objects using adaptive quantization and resampling techniques. J. M. Sotoca 1, F.J. Ferri 1, J. Gutierrez.
Dynamic Bayesian Networks (DBNs)
Computational Stochastic Optimization:
Slide 1 Harnessing Wind in China: Controlling Variability through Location and Regulation DIMACS Workshop: U.S.-China Collaborations in Computer Science.
© 2008 Warren B. Powell Slide 1 The Dynamic Energy Resource Model Warren Powell Alan Lamont Jeffrey Stewart Abraham George © 2007 Warren B. Powell, Princeton.
Universidad de La Habana Lectures 5 & 6 : Difference Equations Kurt Helmes 22 nd September - 2nd October, 2008.
© 2004 Warren B. Powell Slide 1 Outline A car distribution problem.
Approximate Dynamic Programming for High-Dimensional Asset Allocation Ohio State April 16, 2004 Warren Powell CASTLE Laboratory Princeton University
An Optimal Learning Approach to Finding an Outbreak of a Disease Warren Scott Warren Powell
Machine Learning CUNY Graduate Center Lecture 3: Linear Regression.
Efficient Methodologies for Reliability Based Design Optimization
© 2009 Warren B. Powell 1. Optimal Learning for Homeland Security CCICADA Workshop Morgan State, Baltimore, Md. March 7, 2010 Warren Powell With research.
Introduction to Boosting Aristotelis Tsirigos SCLT seminar - NYU Computer Science.
Lecture outline Support vector machines. Support Vector Machines Find a linear hyperplane (decision boundary) that will separate the data.
Planning operation start times for the manufacture of capital products with uncertain processing times and resource constraints D.P. Song, Dr. C.Hicks.
Arizona State University DMML Kernel Methods – Gaussian Processes Presented by Shankar Bhargav.
Advanced Topics in Optimization
Motion Planning in Dynamic Environments Two Challenges for Optimal Path planning.
Slide 1 © 2008 Warren B. Powell Slide 1 Approximate Dynamic Programming for High-Dimensional Problems in Energy Modeling Ohio St. University October 7,
Introducing Information into RM to Model Market Behavior INFORMS 6th RM and Pricing Conference, Columbia University, NY Darius Walczak June 5, 2006.
Equilibrium problems with equilibrium constraints: A new modelling paradigm for revenue management Houyuan Jiang Danny Ralph Stefan Scholtes The Judge.
Walter Hop Web-shop Order Prediction Using Machine Learning Master’s Thesis Computational Economics.
1 CE 530 Molecular Simulation Lecture 7 David A. Kofke Department of Chemical Engineering SUNY Buffalo
1 Statistical Mechanics and Multi- Scale Simulation Methods ChBE Prof. C. Heath Turner Lecture 11 Some materials adapted from Prof. Keith E. Gubbins:
Machine Learning CUNY Graduate Center Lecture 3: Linear Regression.
Computational Stochastic Optimization: Bridging communities October 25, 2012 Warren Powell CASTLE Laboratory Princeton University
Chen Cai, Benjamin Heydecker Presentation for the 4th CREST Open Workshop Operation Research for Software Engineering Methods, London, 2010 Approximate.
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jennifer Rexford Princeton University With Jiayue He, Rui Zhang-Shen, Ying Li,
Chapter 1 Introduction to Simulation
Modeling & Simulation: An Introduction Some slides in this presentation have been copyrighted to Dr. Amr Elmougy.
Optimal Nonlinear Neural Network Controllers for Aircraft Joint University Program Meeting October 10, 2001 Nilesh V. Kulkarni Advisors Prof. Minh Q. Phan.
Computing a posteriori covariance in variational DA I.Gejadze, F.-X. Le Dimet, V.Shutyaev.
An Overview of Dynamic Programming Seminar Series Joe Hartman ISE October 14, 2004.
Stochastic Linear Programming by Series of Monte-Carlo Estimators Leonidas SAKALAUSKAS Institute of Mathematics&Informatics Vilnius, Lithuania
Efficient Integration of Large Stiff Systems of ODEs Using Exponential Integrators M. Tokman, M. Tokman, University of California, Merced 2 hrs 1.5 hrs.
Overview Particle filtering is a sequential Monte Carlo methodology in which the relevant probability distributions are iteratively estimated using the.
Monte Carlo Methods1 T Special Course In Information Science II Tomas Ukkonen
1 S ystems Analysis Laboratory Helsinki University of Technology Kai Virtanen, Tuomas Raivio and Raimo P. Hämäläinen Systems Analysis Laboratory Helsinki.
September Bound Computation for Adaptive Systems V&V Giampiero Campa September 2008 West Virginia University.
© 2007 Warren B. Powell Slide 1 The Dynamic Energy Resource Model Lawrence Livermore National Laboratories September 24, 2007 Warren Powell Alan Lamont.
1 S ystems Analysis Laboratory Helsinki University of Technology Kai Virtanen, Janne Karelahti, Tuomas Raivio, and Raimo P. Hämäläinen Systems Analysis.
VLDB 2006, Seoul1 Indexing For Function Approximation Biswanath Panda Mirek Riedewald, Stephen B. Pope, Johannes Gehrke, L. Paul Chew Cornell University.
Approximate Dynamic Programming and Policy Search: Does anything work? Rutgers Applied Probability Workshop June 6, 2014 Warren B. Powell Daniel R. Jiang.
DaVinci: Dynamically Adaptive Virtual Networks for a Customized Internet Jiayue He, Rui Zhang-Shen, Ying Li, Cheng-Yen Lee, Jennifer Rexford, and Mung.
RSVM: Reduced Support Vector Machines Y.-J. Lee & O. L. Mangasarian First SIAM International Conference on Data Mining Chicago, April 6, 2001 University.
Outline The role of information What is information? Different types of information Controlling information.
CHAPTER 17 O PTIMAL D ESIGN FOR E XPERIMENTAL I NPUTS Organization of chapter in ISSO –Background Motivation Finite sample and asymptotic (continuous)
Wei Sun and KC Chang George Mason University March 2008 Convergence Study of Message Passing In Arbitrary Continuous Bayesian.
INTRO TO OPTIMIZATION MATH-415 Numerical Analysis 1.
DEPARTMENT/SEMESTER ME VII Sem COURSE NAME Operation Research Manav Rachna College of Engg.
Multi-Objective Optimization for Topology Control in Hybrid FSO/RF Networks Jaime Llorca December 8, 2004.
Application of Dynamic Programming to Optimal Learning Problems Peter Frazier Warren Powell Savas Dayanik Department of Operations Research and Financial.
Nonlinear Adaptive Kernel Methods Dec. 1, 2009 Anthony Kuh Chaopin Zhu Nate Kowahl.
Introducing Information into RM to Model Market Behavior INFORMS 6th RM and Pricing Conference, Columbia University, NY Darius Walczak June 5, 2006.
Design and Analysis of Algorithms (09 Credits / 5 hours per week)
Data Transformation: Normalization
LECTURE 09: BAYESIAN ESTIMATION (Cont.)
Boosting and Additive Trees (2)
Clearing the Jungle of Stochastic Optimization
Hidden Markov Models Part 2: Algorithms
ENGG 1801 Engineering Computing
Approximate Dynamic Programming for
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
Support Vector Machines
Ch 3. Linear Models for Regression (2/2) Pattern Recognition and Machine Learning, C. M. Bishop, Previously summarized by Yung-Kyun Noh Updated.
Presentation transcript:

© 2003 Warren B. Powell Slide 1 Approximate Dynamic Programming for High Dimensional Resource Allocation NSF Electric Power workshop November 3, 2003 Warren Powell CASTLE Laboratory Princeton University © 2003 Warren B. Powell, Princeton University

© 2003 Warren B. Powell Slide 2 Schneider National

© 2003 Warren B. Powell Slide 3 Schneider National

© 2003 Warren B. Powell Slide 4

© 2003 Warren B. Powell Slide 5

© 2003 Warren B. Powell Slide 6

© 2003 Warren B. Powell Slide 7 Air Mobility Command Air Mobility Command Fuel Cargo Handling Ramp Space Maintenance Cargo Holding

© 2003 Warren B. Powell Slide 8 The optimization challenge

Special equipment

© 2003 Warren B. Powell Slide 11 State variables Modeling the military airlift problem: »State variables: »Control variables:

© 2003 Warren B. Powell Slide 12 State variables We can formulate the problem of determining what to do with our aircraft as a dynamic program:

© 2003 Warren B. Powell Slide 13 State variables If we only have N=1 aircraft:

© 2003 Warren B. Powell Slide 14 State variables What if we have N>1 aircraft?

© 2003 Warren B. Powell Slide 15 State variables Number of resources Number of attributes Number of zeroes in size of state space

© 2003 Warren B. Powell Slide 16 Outline An algorithmic strategy for high-dimensional asset allocation problems

© 2003 Warren B. Powell Slide 17 Approximate dynamic programming Systems evolve through a cycle of exogenous and endogenous information Time

© 2003 Warren B. Powell Slide 18 Approximate dynamic programming Systems evolve through a cycle of exogenous and endogenous information Time

© 2003 Warren B. Powell Slide 19 Approximate dynamic programming Using this state variable, we obtain the optimality equations: Problem: Curse of dimensionality Three curses State space Outcome space Action space (feasible region)

© 2003 Warren B. Powell Slide 20 Approximate dynamic programming The computational challenge: How do we find ? How do we compute the expectation? How do we find the optimal solution?

© 2003 Warren B. Powell Slide 21 Approximate dynamic programming Approximation methodology: Can’t compute this!!!Don’t know what this is!

© 2003 Warren B. Powell Slide 22 Adaptive dynamic programming Alternative: Change the definition of the state variable: Time

© 2003 Warren B. Powell Slide 23 Adaptive dynamic programming Now our optimality equation looks like: We drop the expectation and solve the conditional problem: Finally, we substitute in our approximation:

© 2003 Warren B. Powell Slide 24 Adaptive dynamic programming Approximating the value function: »We choose approximations of the form:

© 2003 Warren B. Powell Slide 25 Approximate dynamic programming This period Future

© 2003 Warren B. Powell Slide 26 Approximate dynamic programming Our basic strategy: Separable approximation

© 2003 Warren B. Powell Slide 27 Research questions in electric power Special equipment

© 2003 Warren B. Powell Slide 28 Research questions in electric power Two-stage resource allocation under uncertainty

© 2003 Warren B. Powell Slide 29 Approximate dynamic programming

© 2003 Warren B. Powell Slide 30 Approximate dynamic programming

© 2003 Warren B. Powell Slide 31 Approximate dynamic programming

© 2003 Warren B. Powell Slide 32 Approximate dynamic programming We estimate the functions by sampling from our distributions. Marginal value:

© 2003 Warren B. Powell Slide 33 A dynamic network: Approximate dynamic programming t

© 2003 Warren B. Powell Slide 34 Approximate dynamic programming Stepping through time:

© 2003 Warren B. Powell Slide 35 Approximate dynamic programming Iterative learning:

© 2003 Warren B. Powell Slide 36 Nonlinear approximations Number of resources Approximate value function

© 2003 Warren B. Powell Slide 37 Competing algorithmic strategies Competing optimal algorithms: »Discrete dynamic programming Cannot handle even small problems Numerical comparisons are meaningless »Stochastic programming Bender’s decomposition is optimal for this problem class

© 2003 Warren B. Powell Slide 38 Benders decomposition Variations on Bender’s decomposition SPAR algorithm Deterministic approximation Iterations

© 2003 Warren B. Powell Slide 39 Conclusions: »Using sequences of separable, nonlinear approximations conquers the explosive growth with the number of resources. »We are now solving problems with thousands of resources. »But what about the attribute space? Complex equipment and people are typically described by vectors of attributes. We require multidimensional attributes to capture complex assets such as equipment and people. The size of the attribute space grows exponentially in the number of dimensions.

© 2003 Warren B. Powell Slide 40 Benders decomposition Variations on Benders decompositionSPAR Percent over optimal Attribute space = 10 Attribute space = 25 Attribute space = 50 Attribute space = 100

© 2003 Warren B. Powell Slide 41 Benders decomposition Variations on Benders decompositionSPAR Percent over optimal Increasing problem size makes solution much worse With SPAR, the solution gets better.

© 2003 Warren B. Powell Slide 42 Multidimensional attribute spaces decision d

© 2003 Warren B. Powell Slide 43 Multidimensional attribute spaces $450

© 2003 Warren B. Powell Slide 44 NE region PA TX NY Multidimensional attribute spaces

© 2003 Warren B. Powell Slide 45 Hierarchical Aggregation We can use a family of aggregation functions:

We can use different levels of aggregation to capture the value of an asset:

© 2003 Warren B. Powell Slide 47 Hierarchical aggregation Alternative: »Use multiple levels of aggregation at the same time Estimate at gth level of aggregation Weight on gth level of aggregation

© 2003 Warren B. Powell Slide 48 x f(x) Hierarchical aggregation

© 2003 Warren B. Powell Slide 49 x f(x) High structure Moderate structure Zero structure Hierarchical aggregation

© 2003 Warren B. Powell Slide 50 Bayesian weights Weight on disaggregate level Optimal weights Hierarchical aggregation

© 2003 Warren B. Powell Slide 51

© 2003 Warren B. Powell Slide 52 Hierarchical aggregation Aggregate Disaggregate Weighted Combination

© 2003 Warren B. Powell Slide 53 Hierarchical aggregation Iterations Weights Aggregation level 6767 Weight on most disaggregate level Weight on most aggregate levels Optimal weights change as the algorithm progresses:

© 2003 Warren B. Powell Slide 54 Conclusions »Hierarchical aggregation offers a powerful mechanism for handling high dimensional, arbitrary attribute spaces »Combined with the use of separable approximations for handling large numbers of assets, we have a powerful approach for large-scale resource allocation problems.

© 2003 Warren B. Powell Slide 55 Research questions Algorithmic questions: »Stepsizes and rate of convergence We need to improve our understanding of adaptive stepsizes.

© 2003 Warren B. Powell Slide 56 Research questions Algorithmic questions: »Stability: we would like to measure the responsiveness to small changes. The instability limits analyses to big changes. Research: Big simulations will also be unstable. We need to calculate derivatives of simulations. Theory needs to be extended to new problem classes.

© 2003 Warren B. Powell Slide 57 Research questions in electric power Application to electric power: »Fuel optimization (continuous assets): What fuel to purchase when we can switch between fuels Design of fuel contracts Determining prices of forward contracts How much and where to store fuel. »Asset management problems (discrete assets): Unit commitment problems –Control of hydro units Positioning of assets for emergency response –Special equipment –People with specialized training

© 2003 Warren B. Powell Slide 58 Research questions in electric power Special equipment