Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

33

CDC
2008
IEEE

favoriteEmaildiscussreport

206views Control Systems» more CDC 2008»

Approximate dynamic programming using support vector regression

14 years 5 months ago

Approximate dynamic programming using support vector regression

Download web.mit.edu

— This paper presents a new approximate policy iteration algorithm based on support vector regression (SVR). It provides an overview of commonly used cost approximation architectures in approximate dynamic programming problems, explains some difﬁculties encountered by these architectures, and argues that SVR-based architectures can avoid some of these difﬁculties. A key contribution of this paper is to present an extension of the SVR problem to carry out approximate policy iteration by minimizing the Bellman error at selected states. The algorithm does not require trajectory simulations to be performed and is able to utilize a rich set of basis functions in a computationally efﬁcient way. A proof of the algorithm’s correctness in the limit of sampling the entire state space is presented. Finally, computational results for two test problems are shown.

Brett Bethke, Jonathan P. How, Asuman E. Ozdaglar

Real-time Traffic

Approximate Dynamic Programming | Approximate Policy Iteration | CDC 2008 | Control Systems | Policy Iteration Algorithm |

claim paper

Related Content

» Experiments in Value Function Approximation with Sparse Support Vector Regression

» Incorporating prior knowledge in support vector regression

» Model predictive flight control using adaptive support vector regression

» Support Vector Regression Using Mahalanobis Kernels

» Water Demand Prediction using Artificial Neural Networks and Support Vector Regression

» Load Forecasting Using FixedSize Least Squares Support Vector Machines

» Linear parametric noise models for Least Squares Support Vector Machines

» Core Vector Regression for very large regression problems

» Pattern Selection for Support Vector Regression based on Sparseness and Variability

Post Info
More Details (n/a)

Added	29 May 2010
Updated	29 May 2010
Type	Conference
Year	2008
Where	CDC
Authors	Brett Bethke, Jonathan P. How, Asuman E. Ozdaglar

Comments (0)