Protein contacts provide key information for the understanding of protein structure and function, and therefore contact prediction from sequences is an important problem. Recent research shows that some correctly predicted long-range contacts could help topology-level structure modeling. Thus, contact prediction and contact-assisted protein folding also proves the importance of this problem. In this thesis, I will briefly introduce the extant related work, then show how to establish the contact prediction through unsupervised graphical models with topology constraints. Further, I will explain how to use the supervised deep learning methods to further boost the accuracy of contact prediction. Finally, I will propose a scoring system called diversity score to measure the novelty of contact predictions, as well as an algorithm that predicts contacts with respect to the new scoring system.

Emerging applications of machine learning in numerous areas involve continuous gathering of and learning from streams of data. Real-time incorporation of streaming data into the learned models is essential for improved inference in these applications. Further, these applications often involve data that are either inherently gathered at geographically distributed entities or that are intentionally distributed across multiple machines for memory, computational, and/or privacy reasons. Training of models in this distributed, streaming setting requires solving stochastic optimization problems in a collaborative manner over communication links between the physical entities. When the streaming data rate is high compared to the processing capabilities of compute nodes and/or the rate of the communications links, this poses a challenging question: how can one best leverage the incoming data for distributed training under constraints on computing capabilities and/or communications rate? A large body of research has emerged in recent decades to tackle this and related problems. This paper reviews recently developed methods that focus on large-scale distributed stochastic optimization in the compute- and bandwidth-limited regime, with an emphasis on convergence analysis that explicitly accounts for the mismatch between computation, communication and streaming rates. In particular, it focuses on methods that solve: (i) distributed stochastic convex problems, and (ii) distributed principal component analysis, which is a nonconvex problem with geometric structure that permits global convergence. For such methods, the paper discusses recent advances in terms of distributed algorithmic designs when faced with high-rate streaming data. Further, it reviews guarantees underlying these methods, which show there exist regimes in which systems can learn from distributed, streaming data at order-optimal rates.

Multi-view spectral clustering can effectively reveal the intrinsic cluster structure among data by performing clustering on the learned optimal embedding across views. Though demonstrating promising performance in various applications, most of existing methods usually linearly combine a group of pre-specified first-order Laplacian matrices to construct the optimal Laplacian matrix, which may result in limited representation capability and insufficient information exploitation. Also, storing and implementing complex operations on the $n\times n$ Laplacian matrices incurs intensive storage and computation complexity. To address these issues, this paper first proposes a multi-view spectral clustering algorithm that learns a high-order optimal neighborhood Laplacian matrix, and then extends it to the late fusion version for accurate and efficient multi-view clustering. Specifically, our proposed algorithm generates the optimal Laplacian matrix by searching the neighborhood of the linear combination of both the first-order and high-order base Laplacian matrices simultaneously. By this way, the representative capacity of the learned optimal Laplacian matrix is enhanced, which is helpful to better utilize the hidden high-order connection information among data, leading to improved clustering performance. We design an efficient algorithm with proved convergence to solve the resultant optimization problem. Extensive experimental results on nine datasets demonstrate the superiority of our algorithm against state-of-the-art methods, which verifies the effectiveness and advantages of the proposed algorithm.

Secondary structure plays an important role in determining the function of non-coding RNAs. Hence, identifying RNA secondary structures is of great value to research. Computational prediction is a mainstream approach for predicting RNA secondary structure. Unfortunately, even though new methods have been proposed over the past 40 years, the performance of computational prediction methods has stagnated in the last decade. Recently, with the increasing availability of RNA structure data, new methods based on machine-learning technologies, especially deep learning, have alleviated the issue. In this review, we provide a comprehensive overview of RNA secondary structure prediction methods based on machine-learning technologies and a tabularized summary of the most important methods in this field. The current pending issues in the field of RNA secondary structure prediction and future trends are also discussed.

We introduce a new machine-learning-based approach, which we call the Independent Classifier networks (InClass nets) technique, for the nonparameteric estimation of conditional independence mixture models (CIMMs). We approach the estimation of a CIMM as a multi-class classification problem, since dividing the dataset into different categories naturally leads to the estimation of the mixture model. InClass nets consist of multiple independent classifier neural networks (NNs), each of which handles one of the variates of the CIMM. Fitting the CIMM to the data is performed by simultaneously training the individual NNs using suitable cost functions. The ability of NNs to approximate arbitrary functions makes our technique nonparametric. Further leveraging the power of NNs, we allow the conditionally independent variates of the model to be individually high-dimensional, which is the main advantage of our technique over existing non-machine-learning-based approaches. We derive some new results on the nonparametric identifiability of bivariate CIMMs, in the form of a necessary and a (different) sufficient condition for a bivariate CIMM to be identifiable. We provide a public implementation of InClass nets as a Python package called RainDancesVI and validate our InClass nets technique with several worked out examples. Our method also has applications in unsupervised and semi-supervised classification problems.

Continual learning is a branch of deep learning that seeks to strike a balance between learning stability and plasticity. In this paper, we specifically focus on the task-free setting where data are streamed online without task metadata and clear task boundaries. A simple and highly effective algorithm class for this setting is known as Experience Replay (ER) that selectively stores data samples from previous experience and leverages them to interleave memory-based and online batch learning updates. Recent advances in ER have proposed novel methods for scoring which samples to store in memory and which memory samples to interleave with online data during learning updates. In this paper, we contribute a novel Adversarial Shapley value ER (ASER) method that scores memory data samples according to their ability to preserve latent decision boundaries for previously observed classes (to maintain learning stability and avoid forgetting) while interfering with latent decision boundaries of current classes being learned (to encourage plasticity and optimal learning of new class boundaries). Overall, we observe that ASER provides competitive or improved performance on a variety of datasets compared to state-of-the-art ER-based continual learning methods.

Breiman's random forest (RF) can be interpreted as an implicit kernel generator,where the ensuing proximity matrix represents the data-driven RF kernel. Kernel perspective on the RF has been used to develop a principled framework for theoretical investigation of its statistical properties. However, practical utility of the links between kernels and the RF has not been widely explored and systematically evaluated.Focus of our work is investigation of the interplay between kernel methods and the RF. We elucidate the performance and properties of the data driven RF kernels used by regularized linear models in a comprehensive simulation study comprising of continuous, binary and survival targets. We show that for continuous and survival targets, the RF kernels are competitive to RF in higher dimensional scenarios with larger number of noisy features. For the binary target, the RF kernel and RF exhibit comparable performance. As the RF kernel asymptotically converges to the Laplace kernel, we included it in our evaluation. For most simulation setups, the RF and RFkernel outperformed the Laplace kernel. Nevertheless, in some cases the Laplace kernel was competitive, showing its potential value for applications. We also provide the results from real life data sets for the regression, classification and survival to illustrate how these insights may be leveraged in practice.Finally, we discuss further extensions of the RF kernels in the context of interpretable prototype and landmarking classification, regression and survival. We outline future line of research for kernels furnished by Bayesian counterparts of the RF.

Policy gradients methods are a popular and effective choice to train reinforcement learning agents in complex environments. The variance of the stochastic policy gradient is often seen as a key quantity to determine the effectiveness of the algorithm. Baselines are a common addition to reduce the variance of the gradient, but previous works have hardly ever considered other effects baselines may have on the optimization process. Using simple examples, we find that baselines modify the optimization dynamics even when the variance is the same. In certain cases, a baseline with lower variance may even be worse than another with higher variance. Furthermore, we find that the choice of baseline can affect the convergence of natural policy gradient, where certain baselines may lead to convergence to a suboptimal policy for any stepsize. Such behaviour emerges when sampling is constrained to be done using the current policy and we show how decoupling the sampling policy from the current policy guarantees convergence for a much wider range of baselines. More broadly, this work suggests that a more careful treatment of stochasticity in the updates---beyond the immediate variance---is necessary to understand the optimization process of policy gradient algorithms.

We study symmetric spiked matrix models with respect to a general class of noise distributions. Given a rank-1 deformation of a random noise matrix, whose entries are independently distributed with zero mean and unit variance, the goal is to estimate the rank-1 part. For the case of Gaussian noise, the top eigenvector of the given matrix is a widely-studied estimator known to achieve optimal statistical guarantees, e.g., in the sense of the celebrated BBP phase transition. However, this estimator can fail completely for heavy-tailed noise. In this work, we exhibit an estimator that works for heavy-tailed noise up to the BBP threshold that is optimal even for Gaussian noise. We give a non-asymptotic analysis of our estimator which relies only on the variance of each entry remaining constant as the size of the matrix grows: higher moments may grow arbitrarily fast or even fail to exist. Previously, it was only known how to achieve these guarantees if higher-order moments of the noises are bounded by a constant independent of the size of the matrix. Our estimator can be evaluated in polynomial time by counting self-avoiding walks via a color -coding technique. Moreover, we extend our estimator to spiked tensor models and establish analogous results.

As the name says, these approaches rely on the availability of data to extract knowledge and train algorithms. This is opposed to, e.g., modeling approaches in which physiological, physics-based, mathematical, and other equations form the basis of algorithms, or, rule-based systems in which reasoning processes are obtained by translating domain-experts' knowledge into computer-based rules. Focusing on data-driven systems, the data plays a role in several components during the development and actual usage phases. First, we need data to extract knowledge from, i.e., to develop and train algorithms so that they learn-by-example the properties of the problem at hand and get better at solving the problem by repeatedly providing example data. Second, we need to monitor during the development phase how promising the algorithms are and make choices, e.g., concerning optimisation of parameters or choosing different MLparadigms. Methods that don't perform well at all can be discarded, and ones that seem promising can be further optimised. To assess how promising a specific method is, we need to examine how it performs on data that was not used during training. Finally, to objectively assess how well the final'best' system performs, we need to apply completely new data to it that has not been used at all thusfar during the research and development process. Thus, there are at least three stakeholders that have the interest to get as large part of the data pie as possible.