- We investigate experimentally the dynamics of a non-spherical levitated nanoparticle in vacuum. In addition to translation and rotation motion, we observe the light torque-induced precession and nutation of the trapped particle. We provide a theoretical model and numerically simulate the motional frequencies in good agreement with experiments. We measure a torque of $4 \times 10^{-23}$ Nm.
- May 22 2018 stat.AP physics.data-an arXiv:1805.07688v1In this work, we propose a two-stage algorithm based on Bayesian modeling and computation aiming at quantifying analyte concentrations or quantities in complex mixtures with Raman spectroscopy. A hierarchical Bayesian model is built for spectral signal analysis, and reversible-jump Markov chain Monte Carlo (RJMCMC) computation is carried out for model selection and spectral variable estimation. Processing is done in two stages. In the first stage, the peak representations for a target analyte spectrum are learned. In the second, the peak variables learned from the first stage are used to estimate the concentration or quantity of the target analyte in a mixture. Numerical experiments validated its quantification performance over a wide range of simulation conditions and established its advantages for analyte quantification tasks under the small training sample size regime over conventional multivariate regression algorithms. We also used our algorithm to analyze experimental spontaneous Raman spectroscopy data collected for glucose concentration estimation in biopharmaceutical process monitoring applications. Our work shows that this algorithm can be a promising complementary tool alongside conventional multivariate regression algorithms in Raman spectroscopy-based mixture quantification studies, especially when collecting a large training dataset with high quality is challenging or resource-intensive.
- We propose a deep generative Markov State Model (DeepGenMSM) learning framework for inference of metastable dynamical systems and prediction of trajectories. After unsupervised training on time series data, the model contains (i) a probabilistic encoder that maps from high-dimensional configuration space to a small-sized vector indicating the membership to metastable (long-lived) states, (ii) a Markov chain that governs the transitions between metastable states and facilitates analysis of the long-time dynamics, and (iii) a generative part that samples the conditional distribution of configurations in the next time step. The model can be operated in a recursive fashion to generate trajectories to predict the system evolution from a defined starting state and propose new configurations. The DeepGenMSM is demonstrated to provide accurate estimates of the long-time kinetics and generate valid distributions for molecular dynamics (MD) benchmark systems. Remarkably, we show that DeepGenMSMs are able to make long time-steps in molecular configuration space and generate physically realistic structures in regions that were not seen in training data.
- Data is scaling exponentially in fields ranging from genomics to neuroscience to economics. A central question is whether modern machine learning methods can be applied to construct predictive models based on large data sets drawn from complex, natural systems like cells and brains. In machine learning, the predictive power or generalizability of a model is determined by the statistics of training data. In this paper, we ask how predictive inference is impacted when training data is generated by the statistical behavior of a physical system. We develop an information-theoretic analysis of a canonical problem, spin network inference. Our analysis reveals the essential role that thermal fluctuations play in determining the efficiency of predictive inference. Thermal noise drives a system to explore a range of configurations providing `raw' information for a learning algorithm to construct a predictive model. Conversely, thermal energy degrades information by blurring energetic differences between network states. In general, spin networks have an intrinsic optimal temperature at which inference becomes maximally efficient. Simple active learning protocols allow optimization of network temperature, without prior knowledge, to dramatically increase the efficiency of inference. Our results reveal a fundamental link between physics and information and show how the physical environment can be tuned to optimize the efficiency of machine learning.
- May 22 2018 physics.soc-ph arXiv:1805.08063v1Delaunay triangulation can be considered as a type of complex networks. For complex networks, the degree distribution is one of the most important inherent characteristics. In this paper, we first consider the two- and three-dimensional Delaunay Triangulations (DTs) as a type of complex networks, and term it as DT networks. Then we statistically investigate the degree distribution of DT networks. We find that the degree distribution of DT networks well follows the Gaussian distribution in most cases, which differs from the Poisson distribution and the Power-Law distribution for the well-known Small-World networks and Scale-Free networks.
- May 22 2018 physics.bio-ph arXiv:1805.08199v1
- May 22 2018 physics.atom-ph arXiv:1805.08190v1
- May 22 2018 physics.atom-ph arXiv:1805.08185v1
- May 22 2018 physics.optics arXiv:1805.08179v1
- May 22 2018 physics.comp-ph physics.flu-dyn arXiv:1805.08167v1
- May 22 2018 physics.optics quant-ph arXiv:1805.08161v1
- May 22 2018 cond-mat.quant-gas physics.atom-ph arXiv:1805.08128v1
- May 22 2018 physics.flu-dyn arXiv:1805.08088v1
- May 22 2018 physics.soc-ph q-bio.PE arXiv:1805.08046v1
- May 22 2018 physics.optics arXiv:1805.08036v1
- May 22 2018 physics.atom-ph physics.optics arXiv:1805.07986v1
- May 22 2018 physics.optics arXiv:1805.07985v1
- May 22 2018 physics.flu-dyn arXiv:1805.07926v1
- May 22 2018 physics.ins-det arXiv:1805.07922v1
- May 22 2018 physics.flu-dyn arXiv:1805.07893v1
- May 22 2018 physics.optics arXiv:1805.07879v1
- May 22 2018 physics.optics arXiv:1805.07832v1
- May 22 2018 physics.comp-ph physics.flu-dyn arXiv:1805.07801v1
- May 22 2018 physics.soc-ph arXiv:1805.07792v1
- May 22 2018 physics.bio-ph arXiv:1805.07744v1
- May 22 2018 physics.class-ph arXiv:1805.07727v1
- May 22 2018 physics.ins-det hep-ex arXiv:1805.07712v1
- May 22 2018 physics.class-ph arXiv:1805.07711v1
- May 22 2018 physics.optics arXiv:1805.07629v1
- May 22 2018 physics.ed-ph arXiv:1805.07609v1
- May 22 2018 physics.flu-dyn arXiv:1805.07607v1
- May 22 2018 physics.optics arXiv:1805.07558v1