A Multiple Representation Approach to Learning Dynamical Systems

نویسندگان

  • Thomas J. Walsh
  • Michael L. Littman
چکیده

We examine the problem of learning a model of a deterministic dynamical systems from experience. A handful of representation schemes have been proposed for capturing such systems, including POMDPs, PSRs, EPSRs, Diversity, and PSTs. We argue that no single representation should be expected to be ideal in all situations and describe an approach for learning the most succinct representation of an unknown dynamical system.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Dilations for $C^ast$-dynamical systems with abelian groups on Hilbert $C^ast$-modules

‎In this paper we investigate the dilations of completely positive definite representations‎ ‎of (C^ast)-dynamical systems with abelian groups on Hilbert (C^ast)-modules‎. ‎We show that if ((mathcal{A}‎, ‎G,alpha)) is a (C^ast)-dynamical system with (G) an abelian group‎, ‎then every completely positive definite covariant representation ((pi,varphi,E)) of ((mathcal{A}‎, ‎G,alpha)) on a Hilbert ...

متن کامل

Admissible Vectors of a Covariant Representation of a Dynamical System

In this paper, we introduce admissible vectors of covariant representations of a dynamical system which are extensions of the usual ones, and compare them with each other. Also, we give some sufficient conditions for a vector to be admissible vector of a covariant pair of a dynamical system.  In addition, we show the existence of Parseval frames for some special subspaces of $L^2(G)$ related to...

متن کامل

A Dynamical System Approach to Research in Second Language Acquisition

Epistemologically speaking, second language acquisition research (SLAR) might be reconsidered from a complex dynamical system view with interconnected aspects in the ecosystem of language acquisition. The present paper attempts to introduce the tenets of complex system theory and its application in SLAR. It has been suggested that the present dominant traditions in language acquisition research...

متن کامل

Image Classification via Sparse Representation and Subspace Alignment

Image representation is a crucial problem in image processing where there exist many low-level representations of image, i.e., SIFT, HOG and so on. But there is a missing link across low-level and high-level semantic representations. In fact, traditional machine learning approaches, e.g., non-negative matrix factorization, sparse representation and principle component analysis are employed to d...

متن کامل

Designing a quantum genetic controller for tracking the path of quantum systems

Based on learning control methods and computational intelligence, control of quantum systems is an attractive field of study in control engineering. What is important is to establish control approach ensuring that the control process converges to achieve a given control objective and at the same time it is simple and clear. In this paper, a learning control method based on genetic quantum contr...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007