This paper carries out a large dimensional analysis of a variation of kernel ridge regression that we call \emph{centered kernel ridge regression} (CKRR), also known in the literature as kernel ridge regression with offset. This modified technique is obtained by accounting for the bias in the regression problem resulting in the old kernel ridge regression but with \emph{centered} kernels. The analysis is carried out under the assumption that the data is drawn from a Gaussian distribution and heavily relies on tools from random matrix theory (RMT). Under the regime in which the data dimension and the training size grow infinitely large with fixed ratio and under some mild assumptions controlling the data statistics, we show that both the empirical and the prediction risks converge to a deterministic quantities that describe in closed form fashion the performance of CKRR in terms of the data statistics and dimensions. Inspired by this theoretical result, we subsequently build a consistent estimator of the prediction risk based on the training data which allows to optimally tune the design parameters. A key insight of the proposed analysis is the fact that asymptotically a large class of kernels achieve the same minimum prediction risk. This insight is validated with both synthetic and real data.

This paper proposes the use of an optimization algorithm, namely PSO to decide the initial centroids in K-means, to eventually get better accuracy. The vectorized notation of the optimal centroids can be thought of as entities in an optimization space, where the accuracy of K-means over a random subset of the data could act as a fitness measure. The resultant optimal vector can be used as the initial centroids for K-means.

We consider the problem of inferring causal relationships between two or more passively observed variables. While the problem of such causal discovery has been extensively studied especially in the bivariate setting, the majority of current methods assume a linear causal relationship, and the few methods which consider non-linear dependencies usually make the assumption of additive noise. Here, we propose a framework through which we can perform causal discovery in the presence of general non-linear relationships. The proposed method is based on recent progress in non-linear independent component analysis and exploits the non-stationarity of observations in order to recover the underlying sources or latent disturbances. We show rigorously that in the case of bivariate causal discovery, such non-linear ICA can be used to infer the causal direction via a series of independence tests. We further propose an alternative measure of causal direction based on asymptotic approximations to the likelihood ratio, as well as an extension to multivariate causal discovery. We demonstrate the capabilities of the proposed method via a series of simulation studies and conclude with an application to neuroimaging data.

We consider deep networks, trained via stochastic gradient descent to minimize L2 loss, with the training labels perturbed by independent noise at each iteration. We characterize the behavior of the training dynamics near any parameter vector that achieves zero training error, in terms of an implicit regularization term corresponding to the sum over the data points, of the squared L2 norm of the gradient of the model with respect to the parameter vector, evaluated at each data point. We then leverage this general characterization, which holds for networks of any connectivity, width, depth, and choice of activation function, to show that for 2-layer ReLU networks of arbitrary width and L2 loss, when trained on one-dimensional labeled data $(x_1,y_1),\ldots,(x_n,y_n),$ the only stable solutions with zero training error correspond to functions that: 1) are linear over any set of three or more co-linear training points (i.e. the function has no extra "kinks"); and 2) change convexity the minimum number of times that is necessary to fit the training data. Additionally, for 2-layer networks of arbitrary width, with tanh or logistic activations, we show that when trained on a single $d$-dimensional point $(x,y)$ the only stable solutions correspond to networks where the activations of all hidden units at the datapoint, and all weights from the hidden units to the output, take at most two distinct values, or are zero. In this sense, we show that when trained on "simple" data, models corresponding to stable parameters are also "simple"; in short, despite fitting in an over-parameterized regime where the vast majority of expressible functions are complicated and badly behaved, stable parameters reached by training with noise express nearly the "simplest possible" hypothesis consistent with the data. These results shed light on the mystery of why deep networks generalize so well in practice.

The core problem in many Latent Variable Models, widely used in Unsupervised Learning is to find a latent k-simplex K in Rd given perturbed points from it, many of which lie far outside the simplex. This problem was stated in [2] as an open problem. We address this problem under two deterministic assumptions which replace varied stochastic assumptions specific to relevant individual models. Our first contribution is to show that the convex hull K' of the averages of all delta n sized subsets of data points is close to K. We call this subset-smoothing. While K' can have exponentially many vertices, it is easily seen to have a polynomial time Optimization Oracle which in fact runs in time O(nnz(data)). This is the starting point for our algorithm. The algorithm is simple: it has k stages in each of which we use the oracle to find maximum of a carefully chosen linear function over K'; the optimal x is an approximation to a new vertex of K. The simplicity does not carry over to the proof of correctness. The proof is involved and uses existing and new tools from Numerical Analysis, especially angles between singular spaces of close-by matrices. However, the simplicity of the algorithm, especially the fact the only way we use the data is to do matrix-vector products leads to the claimed time bound. This matches the best known algorithms in the special cases and is better when the input is sparse as indeed is the case in many applications. Our algorithm applies to many special cases, including Topic Models, Approximate Non-negative Matrix factorization, Overlapping community Detection and Clustering.

DScribe is a software package for machine learning that provides popular feature transformations ("descriptors") for atomistic materials simulations. DScribe accelerates the application of machine learning for atomistic property prediction by providing user-friendly, off-the-shelf descriptor implementations. The package currently contains implementations for Coulomb matrix, Ewald sum matrix, sine matrix, Many-body Tensor Representation (MBTR), Atom-centered Symmetry Function (ACSF) and Smooth Overlap of Atomic Positions (SOAP). Usage of the package is illustrated for two different applications: formation energy prediction for solids and ionic charge prediction for atoms in organic molecules. The package is freely available under the open-source Apache License 2.0.

Unsupervised relation discovery aims to discover new relations from a given text corpus without annotated data. However, it does not consider existing human annotated knowledge bases even when they are relevant to the relations to be discovered. In this paper, we study the problem of how to use out-of-relation knowledge bases to supervise the discovery of unseen relations, where out-of-relation means that relations to discover from the text corpus and those in knowledge bases are not overlapped. We construct a set of constraints between entity pairs based on the knowledge base embedding and then incorporate constraints into the relation discovery by a variational auto-encoder based algorithm. Experiments show that our new approach can improve the state-of-the-art relation discovery performance by a large margin.

Microbial clades modeling is a challenging problem in biology based on microarray genome sequences, especially in new species gene isolates discovery and category. Marker family genome sequences play important roles in describing specific microbial clades within species, a framework of support vector machine (SVM) based microbial species classification with N-best algorithm is constructed to classify the centroid marker genome fragments randomly generated from marker genome sequences on MetaRef. A time series feature extraction method is proposed by segmenting the centroid gene sequences and mapping into different dimensional spaces. Two ways of data splitting are investigated according to random splitting fragments along genome sequence (DI) , or separating genome sequences into two parts (DII).Two strategies of fragments recognition tasks, dimension-by-dimension and sequence--by--sequence, are investigated. The k-mer size selection, overlap of segmentation and effects of random split percents are also discussed. Experiments on 12390 maker genome sequences belonging to marker families of 17 species from MetaRef show that, both for DI and DII in dimension-by-dimension and sequence-by-sequence recognition, the recognition accuracy rates can achieve above 28\% in top-1 candidate, and above 91\% in top-10 candidate both on training and testing sets overall.

In this work, we study the problem of online optimization of piecewise Lipschitz functions with semi-bandit feedback. This challenging class of non-convex optimization problems often arises in algorithm selection problems for combinatorial settings, where the goal is to find the best algorithm from a large algorithm family for a specific application domain. In these settings, each evaluation of the loss functions in the optimization problem can be computationally expensive, often requiring the learner to run a combinatorial algorithm to measure its performance. Combined with the fact that small differences between similar algorithms in the family can lead to cascading changes in algorithm behavior, efficient online optimization in these settings is a challenging problem. However, we show that in many applications, evaluating the loss function for one algorithm choice can sometimes reveal the loss for a range of similar algorithms, essentially for free. We develop online optimization algorithms capable of using this kind of extra information by working in the semi-bandit feedback setting. Our algorithms achieve regret bounds that are essentially as good as algorithms under full-information feedback and are significantly more computationally efficient. We apply our semi-bandit optimization results to obtain online algorithm selection procedures for two rich families of combinatorial algorithms. We provide the first provable guarantees for online algorithm selection for clustering problems using a family of clustering algorithms containing classic linkage procedures. We also show how to select algorithms from a family of greedy knapsack algorithms with simultaneously lower computational complexity and stronger regret bounds than the best algorithm selection procedures from prior work.

Many of the dynamical processes investigated in the sciences today are characterized by the existence of phenomena on multiple, interconnected time scales that determine the longterm behavior of the process. Examples include the inherently multiscale dynamics of atmospheric vortex-and current formation which needs to be considered for effective weather prediction [24, 29], or the vast difference in time scales on which bounded atomic interactions, side-chain interactions, and the resulting formation of structural motifs occur in biomolecules [19, 12, 10]. An effective approach to analyzing these systems is often the identification of a low-dimensional observable of the system that captures the interesting behaviour on the longest time scale. However, the computerized identification of such observables from simulation data poses a significant computational challenge, especially for high-dimensional systems. Recently, the authors have developed a novel mathematical framework for identifying such essential observables for the slowest time scale of a system [6].