We propose a data aggregation-based algorithm with monotonic convergence to a global optimum for a generalized version of the L1-norm error fitting model with an assumption of the fitting function. Any L1-norm model can be solved optimally using the proposed algorithm if it follows the form of the L1-norm error fitting problem and if the fitting function satisfies the assumption. The proposed algorithm can also solve multi-dimensional fitting problems with arbitrary constraints on the fitting coefficients matrix. The generalized problem includes popular models such as regression, principal component analysis, and the orthogonal Procrustes problem. The results of the computational experiment show that the proposed algorithms are up to 9,000 times faster than the state-of-the-art benchmarks for the problems and datasets studied.

Then we invite you to check out this very friendly introduction we made at Udacity! There are actually 19 videos included in this playlist, covering topics like Linear Regression, Neural Networks, Hierarchical Clustering, and more. Really got the Machine Learning fever? Then consider enrolling in our Machine Learning Nanodegree program. It's the best way to learn everything you need to know to become a successful Machine Learning Engineer!

State-of-the-art algorithms for sparse subspace clustering perform spectral clustering on a similarity matrix typically obtained by representing each data point as a sparse combination of other points using either basis pursuit (BP) or orthogonal matching pursuit (OMP). BP-based methods are often prohibitive in practice while the performance of OMP-based schemes are unsatisfactory, especially in settings where data points are highly similar. In this paper, we propose a novel algorithm that exploits an accelerated variant of orthogonal least-squares to efficiently find the underlying subspaces. We show that under certain conditions the proposed algorithm returns a subspace-preserving solution. Simulation results illustrate that the proposed method compares favorably with BP-based method in terms of running time while being significantly more accurate than OMP-based schemes.

Integrating information from many different data sources to provide better situational awareness is an essential Navy issue. Many data fusion models use statistical methods to reduce statistical errors. Machine learning and big data provide, on the other hand, provides a unique framework for information fusion through our ability to learn what added benefits a different modality can provide. In this work, we provide a novel data fusion method that integrates relational data, provided to us in the form of a graph, and image data. We build an energy model that learns a representation of the data where different data sources are assumed to be similar using a graphical model. The energy model is a non-convex function which we optimize using stochastic gradient descent with momentum. The effectiveness of the model is demonstrated in an automated target recognition example.

Many decisions in healthcare, business, and other policy domains are made without the support of rigorous evidence due to the cost and complexity of performing randomized experiments. Using observational data to answer causal questions is risky: subjects who receive different treatments also differ in other ways that affect outcomes. Many causal inference methods have been developed to mitigate these biases. However, there is no way to know which method might produce the best estimate of a treatment effect in a given study. In analogy to cross-validation, which estimates the prediction error of predictive models applied to a given dataset, we propose synth-validation, a procedure that estimates the estimation error of causal inference methods applied to a given dataset. In synth-validation, we use the observed data to estimate generative distributions with known treatment effects. We apply each causal inference method to datasets sampled from these distributions and compare the effect estimates with the known effects to estimate error. Using simulations, we show that using synth-validation to select a causal inference method for each study lowers the expected estimation error relative to consistently using any single method.

In classification problems, sampling bias between training data and testing data is critical to the ranking performance of classification scores. Such bias can be both unintentionally introduced by data collection and intentionally introduced by the algorithm, such as under-sampling or weighting techniques applied to imbalanced data. When such sampling bias exists, using the raw classification score to rank observations in the testing data can lead to suboptimal results. In this paper, I investigate the optimal calibration strategy in general settings, and develop a practical solution for one specific sampling bias case, where the sampling bias is introduced by stratified sampling. The optimal solution is developed by analytically solving the problem of optimizing the ROC curve. For practical data, I propose a ranking algorithm for general classification models with stratified data. Numerical experiments demonstrate that the proposed algorithm effectively addresses the stratified sampling bias issue. Interestingly, the proposed method shows its potential applicability in two other machine learning areas: unsupervised learning and model ensembling, which can be future research topics.

Convolution Neural Network (CNN) has gained tremendous success in computer vision tasks with its outstanding ability to capture the local latent features. Recently, there has been an increasing interest in extending CNNs to the general spatial domain. Although various types of graph and geometric convolution methods have been proposed, their connections to traditional 2D-convolution are not well-understood. In this paper, we show that depthwise separable convolution is the key to close the gap, based on which we derive a novel Depthwise Separable Graph Convolution that subsumes existing graph convolution methods as special cases of our formulation. Experiments show that the proposed approach consistently outperforms other graph and geometric convolution baselines on benchmark datasets in multiple domains.

Gradient boosted decision trees are a popular machine learning technique, in part because of their ability to give good accuracy with small models. We describe two extensions to the standard tree boosting algorithm designed to increase this advantage. The first improvement extends the boosting formalism from scalar-valued trees to vector-valued trees. This allows individual trees to be used as multiclass classifiers, rather than requiring one tree per class, and drastically reduces the model size required for multiclass problems. We also show that some other popular vector-valued gradient boosted trees modifications fit into this formulation and can be easily obtained in our implementation. The second extension, layer-by-layer boosting, takes smaller steps in function space, which is empirically shown to lead to a faster convergence and to a more compact ensemble. We have added both improvements to the open-source TensorFlow Boosted trees (TFBT) package, and we demonstrate their efficacy on a variety of multiclass datasets. We expect these extensions will be of particular interest to boosted tree applications that require small models, such as embedded devices, applications requiring fast inference, or applications desiring more interpretable models.

Low-rank tensor regression, a new model class that learns high-order correlation from data, has recently received considerable attention. At the same time, Gaussian processes (GP) are well-studied machine learning models for structure learning. In this paper, we demonstrate interesting connections between the two, especially for multi-way data analysis. We show that low-rank tensor regression is essentially learning a multi-linear kernel in Gaussian processes, and the low-rank assumption translates to the constrained Bayesian inference problem. We prove the oracle inequality and derive the average case learning curve for the equivalent GP model. Our finding implies that low-rank tensor regression, though empirically successful, is highly dependent on the eigenvalues of covariance functions as well as variable correlations.

Langevin diffusion is a commonly used tool for sampling from a given distribution. In this work, we establish that when the target density $p^*$ is such that $\log p^*$ is $L$ smooth and $m$ strongly convex, discrete Langevin diffusion produces a distribution $p$ with $KL(p||p^*)\leq \epsilon$ in $\tilde{O}(\frac{d}{\epsilon})$ steps, where $d$ is the dimension of the sample space. We also study the convergence rate when the strong-convexity assumption is absent. By considering the Langevin diffusion as a gradient flow in the space of probability distributions, we obtain an elegant analysis that applies to the stronger property of convergence in KL-divergence and gives a conceptually simpler proof of the best-known convergence results in weaker metrics.