Data science is fast becoming a critical skill for developers and managers across industries, and it looks like a lot of fun as well. But it's pretty complicated - there are a lot of engineering and analytical options to navigate, and it's hard to know if you're doing it right or where the bear traps lie. In this series we explore ways in to making sense of data science - understanding where it's needed and where it's not, and how to make it an asset for you, from people who've been there and done it. This InfoQ article is part of the series "Getting A Handle On Data Science" . You can subscribe to receive notifications via RSS. A lot of Machine Learning (ML) projects consist of fitting a (normally very complicated) function to a dataset with the objective of calculating a number like 1 or 0 (is it spam or not?) for classification problems or a set of numbers (e.g., weekly sales of a product) for regression ones.

This algorithm forms the basis for many modern day ML algorithms, most notably neural networks. In addition, we'll discuss the perceptron algorithm's cousin, logistic regression. And then we'll conclude with an introduction to SVMs, or support vector machines, which are perhaps one of the most flexible algorithms used today. In machine learning, there are two general classes of algorithms. You'll remember that in our last post we discussed regression and classification.

To model categorical response variables given their covariates, we propose a permuted and augmented stick-breaking (paSB) construction that one-to-one maps the observed categories to randomly permuted latent sticks. This new construction transforms multinomial regression into regression analysis of stick-specific binary random variables that are mutually independent given their covariate-dependent stick success probabilities, which are parameterized by the regression coefficients of their corresponding categories. The paSB construction allows transforming an arbitrary cross-entropy-loss binary classifier into a Bayesian multinomial one. Specifically, we parameterize the negative logarithms of the stick failure probabilities with a family of covariate-dependent softplus functions to construct nonparametric Bayesian multinomial softplus regression, and transform Bayesian support vector machine (SVM) into Bayesian multinomial SVM. These Bayesian multinomial regression models are not only capable of providing probability estimates, quantifying uncertainty, and producing nonlinear classification decision boundaries, but also amenable to posterior simulation. Example results demonstrate their attractive properties and appealing performance.

Sparse Bayesian learning is one of the state-of- the-art machine learning algorithms, which is able to make stable and reliable probabilistic predictions. However, some of these algorithms, e.g. probabilistic classification vector machine (PCVM) and relevant vector machine (RVM), are not capable of eliminating irrelevant and redundant features which could lead to performance degradation. To tackle this problem, in this paper, we propose a sparse Bayesian classifier which simultaneously selects the relevant samples and features. We name this classifier a probabilistic feature selection and classification vector machine (PFCVM), in which truncated Gaussian distributions are em- ployed as both sample and feature priors. In order to derive the analytical solution for the proposed algorithm, we use Laplace approximation to calculate approximate posteriors and marginal likelihoods. Finally, we obtain the optimized parameters and hyperparameters by the type-II maximum likelihood method. The experiments on synthetic data set, benchmark data sets and high dimensional data sets validate the performance of PFCVM under two criteria: accuracy of classification and efficacy of selected features. Finally, we analyze the generalization performance of PFCVM and derive a generalization error bound for PFCVM. Then by tightening the bound, we demonstrate the significance of the sparseness for the model.

A new type of End-to-End system for text-dependent speaker verification is presented in this paper. Previously, using the phonetically discriminative/speaker discriminative DNNs as feature extractors for speaker verification has shown promising results. The extracted frame-level (DNN bottleneck, posterior or d-vector) features are equally weighted and aggregated to compute an utterance-level speaker representation (d-vector or i-vector). In this work we use speaker discriminative CNNs to extract the noise-robust frame-level features. These features are smartly combined to form an utterance-level speaker vector through an attention mechanism. The proposed attention model takes the speaker discriminative information and the phonetic information to learn the weights. The whole system, including the CNN and attention model, is joint optimized using an end-to-end criterion. The training algorithm imitates exactly the evaluation process --- directly mapping a test utterance and a few target speaker utterances into a single verification score. The algorithm can automatically select the most similar impostor for each target speaker to train the network. We demonstrated the effectiveness of the proposed end-to-end system on Windows $10$ "Hey Cortana" speaker verification task.

We propose a method for recovering the structure of a sparse undirected graphical model when very few samples are available. The method decides about the presence or absence of bonds between pairs of variable by considering one pair at a time and using a closed form formula, analytically derived by calculating the posterior probability for every possible model explaining a two body system using Jeffreys prior. The approach does not rely on the optimisation of any cost functions and consequently is much faster than existing algorithms. Despite this time and computational advantage, numerical results show that for several sparse topologies the algorithm is comparable to the best existing algorithms, and is more accurate in the presence of hidden variables. We apply this approach to the analysis of US stock market data and to neural data, in order to show its efficiency in recovering robust statistical dependencies in real data with non stationary correlations in time and space.

We propose a methodology for testing linear hypothesis in high-dimensional linear models. The proposed test does not impose any restriction on the size of the model, i.e. model sparsity or the loading vector representing the hypothesis. Providing asymptotically valid methods for testing general linear functions of the regression parameters in high-dimensions is extremely challenging -- especially without making restrictive or unverifiable assumptions on the number of non-zero elements. We propose to test the moment conditions related to the newly designed restructured regression, where the inputs are transformed and augmented features. These new features incorporate the structure of the null hypothesis directly. The test statistics are constructed in such a way that lack of sparsity in the original model parameter does not present a problem for the theoretical justification of our procedures. We establish asymptotically exact control on Type I error without imposing any sparsity assumptions on model parameter or the vector representing the linear hypothesis. Our method is also shown to achieve certain optimality in detecting deviations from the null hypothesis. We demonstrate the favorable finite-sample performance of the proposed methods, via a number of numerical and a real data example.

Autoregression is a time series model that uses observations from previous time steps as input to a regression equation to predict the value at the next time step. It is a very simple idea that can result in accurate forecasts on a range of time series problems. In this tutorial, you will discover how to implement an autoregressive model for time series forecasting with Python. Autoregression Models for Time Series Forecasting With Python Photo by Umberto Salvagnin, some rights reserved. A regression model, such as linear regression, models an output value based on a linear combination of input values.

Based on different understandings of \( y_i \) we can have different problems, such as regression, classification, ordering, etc. We need to find a way to find the best parameters given the training data. In order to do so, we need to define a so-called objective function, to measure the performance of the model given a certain set of parameters. A very important fact about objective functions is they must always contain two parts: training loss and regularization. The training loss measures how predictive our model is on training data.

Can we recover a complex signal from its Fourier magnitudes? More generally, given a set of $m$ measurements, $y_k = |\mathbf a_k^* \mathbf x|$ for $k = 1, \dots, m$, is it possible to recover $\mathbf x \in \mathbb{C}^n$ (i.e., length-$n$ complex vector)? This **generalized phase retrieval** (GPR) problem is a fundamental task in various disciplines, and has been the subject of much recent investigation. Natural nonconvex heuristics often work remarkably well for GPR in practice, but lack clear theoretical explanations. In this paper, we take a step towards bridging this gap. We prove that when the measurement vectors $\mathbf a_k$'s are generic (i.i.d. complex Gaussian) and the number of measurements is large enough ($m \ge C n \log^3 n$), with high probability, a natural least-squares formulation for GPR has the following benign geometric structure: (1) there are no spurious local minimizers, and all global minimizers are equal to the target signal $\mathbf x$, up to a global phase; and (2) the objective function has a negative curvature around each saddle point. This structure allows a number of iterative optimization methods to efficiently find a global minimizer, without special initialization. To corroborate the claim, we describe and analyze a second-order trust-region algorithm.