%0 Journal Article %1 ge2018simulated %A Ge, Rong %A Lee, Holden %A Risteski, Andrej %D 2018 %K bayesian dynamic optimization readings stats %T Simulated Tempering Langevin Monte Carlo II: An Improved Proof using Soft Markov Chain Decomposition %U http://arxiv.org/abs/1812.00793 %X A key task in Bayesian machine learning is sampling from distributions that are only specified up to a partition function (i.e., constant of proportionality). One prevalent example of this is sampling posteriors in parametric distributions, such as latent-variable generative models. However sampling (even very approximately) can be #P-hard. Classical results going back to Bakry and Émery (1985) on sampling focus on log-concave distributions, and show a natural Markov chain called Langevin diffusion mixes in polynomial time. However, all log-concave distributions are uni-modal, while in practice it is very common for the distribution of interest to have multiple modes. In this case, Langevin diffusion suffers from torpid mixing. We address this problem by combining Langevin diffusion with simulated tempering. The result is a Markov chain that mixes more rapidly by transitioning between different temperatures of the distribution. We analyze this Markov chain for a mixture of (strongly) log-concave distributions of the same shape. In particular, our technique applies to the canonical multi-modal distribution: a mixture of gaussians (of equal variance). Our algorithm efficiently samples from these distributions given only access to the gradient of the log-pdf. For the analysis, we introduce novel techniques for proving spectral gaps based on decomposing the action of the generator of the diffusion. Previous approaches rely on decomposing the state space as a partition of sets, while our approach can be thought of as decomposing the stationary measure as a mixture of distributions (a "soft partition"). Additional materials for the paper can be found at http://tiny.cc/glr17. The proof and results have been improved and generalized from the precursor at www.arxiv.org/abs/1710.02736.

@article{ge2018simulated, abstract = {A key task in Bayesian machine learning is sampling from distributions that are only specified up to a partition function (i.e., constant of proportionality). One prevalent example of this is sampling posteriors in parametric distributions, such as latent-variable generative models. However sampling (even very approximately) can be #P-hard. Classical results going back to Bakry and \'Emery (1985) on sampling focus on log-concave distributions, and show a natural Markov chain called Langevin diffusion mixes in polynomial time. However, all log-concave distributions are uni-modal, while in practice it is very common for the distribution of interest to have multiple modes. In this case, Langevin diffusion suffers from torpid mixing. We address this problem by combining Langevin diffusion with simulated tempering. The result is a Markov chain that mixes more rapidly by transitioning between different temperatures of the distribution. We analyze this Markov chain for a mixture of (strongly) log-concave distributions of the same shape. In particular, our technique applies to the canonical multi-modal distribution: a mixture of gaussians (of equal variance). Our algorithm efficiently samples from these distributions given only access to the gradient of the log-pdf. For the analysis, we introduce novel techniques for proving spectral gaps based on decomposing the action of the generator of the diffusion. Previous approaches rely on decomposing the state space as a partition of sets, while our approach can be thought of as decomposing the stationary measure as a mixture of distributions (a "soft partition"). Additional materials for the paper can be found at http://tiny.cc/glr17. The proof and results have been improved and generalized from the precursor at www.arxiv.org/abs/1710.02736.}, added-at = {2020-02-20T15:51:15.000+0100}, author = {Ge, Rong and Lee, Holden and Risteski, Andrej}, biburl = {https://www.bibsonomy.org/bibtex/2ae85f39f5f30fe1d9596e5e89c246460/kirk86}, description = {[1812.00793] Simulated Tempering Langevin Monte Carlo II: An Improved Proof using Soft Markov Chain Decomposition}, interhash = {da67941154ac4ff6cd0ebde2bd3fd111}, intrahash = {ae85f39f5f30fe1d9596e5e89c246460}, keywords = {bayesian dynamic optimization readings stats}, note = {cite arxiv:1812.00793Comment: 69 pages. arXiv admin note: text overlap with arXiv:1710.02736}, timestamp = {2020-02-20T15:51:15.000+0100}, title = {Simulated Tempering Langevin Monte Carlo II: An Improved Proof using Soft Markov Chain Decomposition}, url = {http://arxiv.org/abs/1812.00793}, year = 2018 }