<ul data-eligibleForWebStory="false"><li>Sequential Monte Carlo Squared (SMC$^2$) method is designed to improve accuracy and computational efficiency in Bayesian inference.</li><li>The incorporation of second-order proposals, utilizing Hessian information, within SMC$^2$ framework enhances exploration of the posterior distribution.</li><li>Experimental results on synthetic models show the advantages of using second-order proposals in terms of step-size selection and posterior approximation accuracy.</li><li>This study extends the Metropolis-Adjusted Langevin Algorithm (MALA) by incorporating second-order information for better exploration of probability regions.</li></ul>

Hess-MC2: Sequential Monte Carlo Squared using Hessian Information and Second Order Proposals

Discover more