MULTI-INSTRUMENT DETECTION IN POLYPHONIC MUSIC USING GAUSSIAN MIXTURE BASED FACTORIAL HMM

We formulate the problem of detecting the constituent instruments in a polyphonic music piece as a joint decoding problem. From monophonic data, parametric Gaussian Mixture Hidden Markov Models (GM-HMM) are obtained for each instrument. We propose a method to use the above models in a factorial framework, termed as Factorial GM-HMM (F-GM-HMM). The states are jointly inferred to explain the evolution of each instrument in the mixture observation sequence. The dependencies are decoupled using variational inference technique. We show that the joint time evolution of all instruments' states can be captured using F-GM-HMM. We compare performance of proposed method with that of Student's-t mixture model (tMM) and GM-HMM in an existing latent variable framework. Experiments on two to five polyphony with 8 instrument models trained on the RWC dataset, tested on RWC and TRIOS datasets show that F-GM-HMM gives an advantage over the other considered models in segments containing co-occurring instruments.

Asymptotic behavior and interpretation of virtual states: The effects of confinement and of basis sets

A real-space high order finite difference method is used to analyze the effect of spherical domain size on the Hartree-Fock (and density functional theory) virtual eigenstates. We show the domain size dependence of both positive and negative virtual eigenvalues of the Hartree-Fock equations for small molecules. We demonstrate that positive states behave like a particle in spherical well and show how they approach zero. For the negative eigenstates, we show that large domains are needed to get the correct eigenvalues. We compare our results to those of Gaussian basis sets and draw some conclusions for real-space, basis-sets, and plane-waves calculations. (C) 2016 AIP Publishing LLC.

Strength of Offshore Pipelines in Different Operation States and Analyzing Software

在近海管线的铺设、安装、使用过程中有多种作业状态：在位、悬跨、挖沟、提吊、铺管等。各种状态下管线的受力特点不同,加上管线结构、海况和海底土壤等因素又都很复杂,所以近海管线的强度分析难度大、内容多。分别采用解析方法、数值方法（有限元法、打靶法）和二者结合来解决理论上的（如几何非线性、动边界等）、实用性方面的难点。在理论分析的基础上,编制了符合产业部门工程师使用要求的近海管线强度分析软件。介绍了该软件进行力学分析时采用的理论以及软件界面。

Bayesian model selection for time series using Markov chain Monte Carlo

We present a stochastic simulation technique for subset selection in time series models, based on the use of indicator variables with the Gibbs sampler within a hierarchical Bayesian framework. As an example, the method is applied to the selection of subset linear AR models, in which only significant lags are included. Joint sampling of the indicators and parameters is found to speed convergence. We discuss the possibility of model mixing where the model is not well determined by the data, and the extension of the approach to include non-linear model terms.

Population of nonnative states of lysozyme variants drives amyloid fibril formation.

The propensity of protein molecules to self-assemble into highly ordered, fibrillar aggregates lies at the heart of understanding many disorders ranging from Alzheimer's disease to systemic lysozyme amyloidosis. In this paper we use highly accurate kinetic measurements of amyloid fibril growth in combination with spectroscopic tools to quantify the effect of modifications in solution conditions and in the amino acid sequence of human lysozyme on its propensity to form amyloid fibrils under acidic conditions. We elucidate and quantify the correlation between the rate of amyloid growth and the population of nonnative states, and we show that changes in amyloidogenicity are almost entirely due to alterations in the stability of the native state, while other regions of the global free-energy surface remain largely unmodified. These results provide insight into the complex dynamics of a macromolecule on a multidimensional energy landscape and point the way for a better understanding of amyloid diseases.

Parameter Estimation for Hidden Markov Models with Intractable Likelihoods

Approximate Bayesian computation (ABC) is a popular technique for analysing data for complex models where the likelihood function is intractable. It involves using simulation from the model to approximate the likelihood, with this approximate likelihood then being used to construct an approximate posterior. In this paper, we consider methods that estimate the parameters by maximizing the approximate likelihood used in ABC. We give a theoretical analysis of the asymptotic properties of the resulting estimator. In particular, we derive results analogous to those of consistency and asymptotic normality for standard maximum likelihood estimation. We also discuss how sequential Monte Carlo methods provide a natural method for implementing our likelihood-based ABC procedures.

Bayesian learning of noisy Markov decision processes

This work addresses the problem of estimating the optimal value function in a Markov Decision Process from observed state-action pairs. We adopt a Bayesian approach to inference, which allows both the model to be estimated and predictions about actions to be made in a unified framework, providing a principled approach to mimicry of a controller on the basis of observed data. A new Markov chain Monte Carlo (MCMC) sampler is devised for simulation from theposterior distribution over the optimal value function. This step includes a parameter expansion step, which is shown to be essential for good convergence properties of the MCMC sampler. As an illustration, the method is applied to learning a human controller.

Limiting States Of Internal Wave Rays

Existing models of baroclinic tides are based upon the "traditional approximation'', i. e., neglect of the horizontal component of the Earth's rotation, leading to a well- known conclusion that no freely propagating internal waves can exist beyond the critical latitude and the wave rays are symmetric to the vertical. However, recent studies have contended that the situation may change if both the vertical and horizontal components of the Earth's rotation are taken into account. With the full account of the Coriolis force, characteristics of the internal wavefield generated by tidal flow over uneven topography are investigated. It is found that "nontraditional effects'' profoundly change not only the dynamics of internal waves but also the rate at which the barotropic tidal energy is fed into the internal wavefield. Discarding the traditional approximation, internal waves are proved to be able to generate poleward of the critical latitude, rays of which are no longer symmetric and the limiting values of ray angles become greater or less than 90 degrees, depending on the local latitude and the direction of ray. More importantly, in contrast to the predictions of models based upon the traditional approximation, a substantial conversion occurs in the situations when stratification is so weak that the buoyancy frequency is below the tidal one.