اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدAdaptive MCMC via Combining Local Samplers
86
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Markov chain Monte Carlo (MCMC) methods are widely used in machine learning. One of the major problems with MCMC is the question of how to design chains that mix fast over the whole state space; in particular, how to select the parameters of an MCMC algorithm. Here we take a different approach and, similarly to parallel MCMC methods, instead of trying to find a single chain that samples from the whole distribution, we combine samples from several chains run in parallel, each exploring only parts of the state space (e.g., a few modes only). The chains are prioritized based on kernel Stein discrepancy, which provides a good measure of performance locally. The samples from the independent chains are combined using a novel technique for estimating the probability of different regions of the sample space. Experimental results demonstrate that the proposed algorithm may provide significant speedups in different sampling problems. Most importantly, when combined with the state-of-the-art NUTS algorithm as the base MCMC sampler, our method remained competitive with NUTS on sampling from unimodal distributions, while significantly outperforming state-of-the-art competitors on synthetic multimodal problems as well as on a challenging sensor localization task.
قيم البحث
اقرأ أيضاً
Monte Carlo methods are the standard procedure for estimating complicated integrals of multidimensional Bayesian posterior distributions. In this work, we focus on LAIS, a class of adaptive importance samplers where Markov chain Monte Carlo (MCMC) al
gorithms are employed to drive an underlying multiple importance sampling (IS) scheme. Its power lies in the simplicity of the layered framework: the upper layer locates proposal densities by means of MCMC algorithms; while the lower layer handles the multiple IS scheme, in order to compute the final estimators. The modular nature of LAIS allows for different possible choices in the upper and lower layers, that will have different performance and computational costs. In this work, we propose different enhancements in order to increase the efficiency and reduce the computational cost, of both upper and lower layers. The different variants are essential if we aim to address computational challenges arising in real-world applications, such as highly concentrated posterior distributions (due to large amounts of data, etc.). Hamiltonian-driven importance samplers are presented and tested. Furthermore, we introduce different strategies for designing cheaper schemes, for instance, recycling samples generated in the upper layer and using them in the final estimators in the lower layer. Numerical experiments show the benefits of the proposed schemes as compared to the vanilla version of LAIS and other benchmark methods.
Component-wise MCMC algorithms, including Gibbs and conditional Metropolis-Hastings samplers, are commonly used for sampling from multivariate probability distributions. A long-standing question regarding Gibbs algorithms is whether a deterministic-s
can (systematic-scan) sampler converges faster than its random-scan counterpart. We answer this question when the samplers involve two components by establishing an exact quantitative relationship between the $L^2$ convergence rates of the two samplers. The relationship shows that the deterministic-scan sampler converges faster. We also establish qualitative relations among the convergence rates of two-component Gibbs samplers and some conditional Metropolis-Hastings variants. For instance, it is shown that if some two-component conditional Metropolis-Hastings samplers are geometrically ergodic, then so are the associated Gibbs samplers.
We consider the problem of designing an adaptive sequence of questions that optimally classify a candidates ability into one of several categories or discriminative grades. A candidates ability is modeled as an unknown parameter, which, together with
the difficulty of the question asked, determines the likelihood with which s/he is able to answer a question correctly. The learning algorithm is only able to observe these noisy responses to its queries. We consider this problem from a fixed confidence-based $delta$-correct framework, that in our setting seeks to arrive at the correct ability discrimination at the fastest possible rate while guaranteeing that the probability of error is less than a pre-specified and small $delta$. In this setting we develop lower bounds on any sequential questioning strategy and develop geometrical insights into the problem structure both from primal and dual formulation. In addition, we arrive at algorithms that essentially match these lower bounds. Our key conclusions are that, asymptotically, any candidate needs to be asked questions at most at two (candidate ability-specific) levels, although, in a reasonably general framework, questions need to be asked only at a single level. Further, and interestingly, the problem structure facilitates endogenous exploration, so there is no need for a separately designed exploration stage in the algorithm.
In this paper we provide performance guarantees for hypocoercive non-reversible MCMC samplers $X_t$ with invariant measure $mu^*$ and our results apply in particular to the Langevin equation, Hamiltonian Monte-Carlo, and the bouncy particle and zig-z
ag samplers. Specifically, we establish a concentration inequality of Bernstein type for ergodic averages $frac{1}{T} int_0^T f(X_t), dt$. As a consequence we provide performance guarantees: (a) explicit non-asymptotic confidence intervals for $int f dmu^*$ when using a finite time ergodic average with given initial condition $mu$ and (b) uncertainty quantification bounds, expressed in terms of relative entropy rate, on the bias of $int f dmu^*$ when using an alternative or approximate processes $widetilde{X}_t$. (Results in (b) generalize recent results (arXiv:1812.05174) from the authors for coercive dynamics.) The concentration inequality is proved by combining the approach via Feynmann-Kac semigroups first noted by Wu with the hypocoercive estimates of Dolbeault, Mouhot and Schmeiser (arXiv:1005.1495) developed for the Langevin equation and recently generalized to partially deterministic Markov processes by Andrieu et al. (arXiv:1808.08592)
In this work, we propose a multi-agent actor-critic reinforcement learning (RL) algorithm to accelerate the multi-level Monte Carlo Markov Chain (MCMC) sampling algorithms. The policies (actors) of the agents are used to generate the proposal in the
MCMC steps; and the critic, which is centralized, is in charge of estimating the long term reward. We verify our proposed algorithm by solving an inverse problem with multiple scales. There are several difficulties in the implementation of this problem by using traditional MCMC sampling. Firstly, the computation of the posterior distribution involves evaluating the forward solver, which is very time consuming for a problem with heterogeneous. We hence propose to use the multi-level algorithm. More precisely, we use the generalized multiscale finite element method (GMsFEM) as the forward solver in evaluating a posterior distribution in the multi-level rejection procedure. Secondly, it is hard to find a function which can generate samplings which are meaningful. To solve this issue, we learn an RL policy as the proposal generator. Our experiments show that the proposed method significantly improves the sampling process
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
