Abstract
This paper addresses the problem of sensitivity analysis for finite-horizon performance measures of general Markov chains. We derive closed-form expressions and associated unbiased gradient estimators for the derivatives of finite products of Markov kernels by measure-valued differentiation (MVD). In the MVD setting, the derivatives of Markov kernels, called
$\mathcal{D}$
-derivatives, are defined with respect to a class of performance functions
$\mathcal{D}$
such that, for any performance measure
$g\in\mathcal{D}$
, the derivative of the integral of g with respect to the one-step transition probability of the Markov chain exists. The MVD approach (i) yields results that can be applied to performance functions out of a predefined class, (ii) allows for a product rule of differentiation, that is, analyzing the derivative of the transition kernel immediately yields finite-horizon results, (iii) provides an operator language approach to the differentiation of Markov chains and (iv) clearly identifies the trade-off between the generality of the performance classes that can be analyzed and the generality of the classes of measures (Markov kernels). The
$\mathcal{D}$
-derivative of a measure can be interpreted in terms of various (unbiased) gradient estimators and the product rule for
$\mathcal {D}$
-differentiation yields a product-rule for various gradient estimators.
[1]
J. M. Hammersley,et al.
Conditional Monte Carlo
,
1956,
JACM.
[2]
P. Meyer,et al.
Probabilities and potential C
,
1978
.
[3]
Paul Glasserman,et al.
Gradient Estimation Via Perturbation Analysis
,
1990
.
[4]
Xi-Ren Cao,et al.
Perturbation analysis of discrete event dynamic systems
,
1991
.
[5]
P. Glasserman,et al.
Some Guidelines and Guarantees for Common Random Numbers
,
1992
.
[6]
Xi-Ren Cao,et al.
Realization Probabilities: The Dynamics of Queuing Systems
,
1994
.
[7]
George Ch. Pflug,et al.
Optimization of Stochastic Models
,
1996
.
[8]
Arie Hordijk,et al.
Derivatives of Markov Kernels and Their Jordan Decomposition
,
2003
.
[9]
Arie Hordijk,et al.
Single-run gradient estimation via measure-valued differentiation
,
2004,
IEEE Transactions on Automatic Control.
[10]
F. Vázquez-Abad,et al.
Measure valued differentiation for random horizon problems
,
2006
.