Hamiltonian Monte Carlo methods for Subset Simulation in reliability analysis

@article{Wang2019HamiltonianMC,
  title={Hamiltonian Monte Carlo methods for Subset Simulation in reliability analysis},
  author={Ziqi Wang and Marco Broccardo and Junho Song},
  journal={Structural Safety},
  year={2019}
}
A DIRECT HAMILTONIAN MCMC APPROACH FOR RELIABILITY ESTIMATION
  • Hamed Nikbakht, K. Papakonstantinou
  • Computer Science
    Proceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECOMP 2019)
  • 2019
TLDR
This work introduces a gradient-based Hamiltonian Markov Chain Monte Carlo framework, termed Approximate Sampling Target with Post-processing Adjustment (ASTPA), to construct a relevant target distribution by weighting the high-dimensional random variable space through a one-dimensional likelihood model, using the limit-state function.
Implementing the Hamiltonian Monte Carlo Sampling Algorithm in Stochastic Assessment of Power Systems
TLDR
It is investigated whether the proposal is able to mitigate adverse effects of the standard Metropolis–Hastings sampling algorithm, such as random-walk, low acceptance rates and slow convergence, so that an efficient and robust sampler can be considered for stochastic studies of power systems.
Estimation of Failure Probabilities via Local Subset Approximations
TLDR
The subset simulation method which approaches a failure event using a decreasing sequence of nested intermediate failure events is considered, and the partial least squares regression, a gradient-free reduction method, is employed locally to explore and utilize a low-dimensional subspace within a Markov chain.
Estimation of Failure Probabilities via Local Subset Approximations
TLDR
The subset simulation method which approaches a failure event using a decreasing sequence of nested intermediate failure events is considered, and the partial least squares (PLS) regression, a gradient-free reduction method, is employed locally to explore and utilize a low-dimensional subspace within a Markov chain.
Hamiltonian MCMC methods for estimating rare events probabilities in high-dimensional problems
TLDR
The focus in this work is on precisely quantifying these probabilities, often encountered in reliability analysis of complex engineering systems, based on an introduced framework termed Approximate Sampling Target with Post-processing Adjustment (ASTPA), which is integrated with and supported by gradient-based Hamiltonian Markov Chain Monte Carlo (HMCMC) methods.
An adaptive subset simulation algorithm for system reliability analysis with discontinuous limit states
TLDR
An adaptive subset simulation algorithm to determine the reliability of systems with discontinuous limit state functions is proposed, which chooses the number of samples and the conditional probability adaptively.
Reliability Analysis Based on Optimization Random Forest Model and MCMC
  • Fan Yang, J. Ren
  • Engineering, Computer Science
    Computer Modeling in Engineering & Sciences
  • 2020
TLDR
A novel method of reliability analysis combining Monte Carlo Markov Chain (MCMC) with random forest algorithm was proposed and examples demonstrate the proposed method possesses higher computational efficiency and accuracy.
Stochastic Methods for Emulation, Calibration and Reliability Analysis of Engineering Models
TLDR
New MCMC algorithms are presented which allow the use of full Bayesian emulators by sampling from their respective multimodal posteriors, and the GP surrogate model’s probabilistic statements are exploited and the data assimilation process is improved.
...
...

References

SHOWING 1-10 OF 49 REFERENCES
On MCMC algorithm for Subset Simulation
  • S. Au
  • Computer Science
  • 2016
Probabilistic Inference Using Markov Chain Monte Carlo Methods
TLDR
The role of probabilistic inference in artificial intelligence is outlined, the theory of Markov chains is presented, and various Markov chain Monte Carlo algorithms are described, along with a number of supporting techniques.
The No-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo
TLDR
The No-U-Turn Sampler (NUTS), an extension to HMC that eliminates the need to set a number of steps L, and derives a method for adapting the step size parameter {\epsilon} on the fly based on primal-dual averaging.
Bayesian Model Updating Using Hybrid Monte Carlo Simulation with Application to Structural Dynamic Models with Many Uncertain Parameters
In recent years, Bayesian model updating techniques based on measured data have been applied to system identification of structures and to structural health monitoring. A fully probabilistic Bayesian
An adaptive Metropolis algorithm
TLDR
An adaptive Metropolis (AM) algorithm, where the Gaussian proposal distribution is updated along the process using the full information cumulated so far, which establishes here that it has the correct ergodic properties.
Annealed importance sampling
TLDR
It is shown how one can use the Markov chain transitions for such an annealing sequence to define an importance sampler, which can be seen as a generalization of a recently-proposed variant of sequential importance sampling.
...
...