H. Okamura, T. Sakoh, and T. Dohi (Japan)
Software Reliability, Parameter Estimation, Varia tional Bayesian Approach, Exponential Model, Non Homogeneous Poisson Process
In this paper we develop a novel parameter estimation al gorithm for the well-known exponential software reliabil ity model based on the non-homogeneous Poisson process. More specifically, an approximate solution algorithm is provided within a framework of variational Bayesian ap proach, where the Bayesian parameter estimation can be drastically improved in terms of computation effort. In a numerical example, we show the effectiveness of the proposed algorithm and investigate its approximate perfor mance by comparing with the exact solution.
Important Links:
Go Back
