Hessian corrections to the Metropolis Adjusted Langevin Algorithm

2015/07/22 by Thomas House, House, Thomas
Mathematics · Biochemistry, Genetics and Molecular Biology · Physics and Astronomy · #Markov Chains and Monte Carlo Methods #Protein Structure and Dynamics #Theoretical and Computational Physics

paper · pdf · doi:10.48550/arxiv.1507.06336

Abstract

A natural method for the introduction of second-order derivatives of the log likelihood into MCMC algorithms is introduced, based on Taylor expansion of the Langevin equation followed by exact solution of the truncated system.

Citations

• Markov Chain Monte Carlo in Practice: A Roundtable Discussion

Related

• Hessian corrections to Hybrid Monte Carlo
• Underdamped Langevin MCMC: A non-asymptotic analysis
• Metropolis-adjusted Subdifferential Langevin Algorithm
• The shifted ODE method for underdamped Langevin MCMC
• Hess-MC2: Sequential Monte Carlo Squared using Hessian Information and Second Order Proposals