The use of electronic cigarette (e-cigarette) is increasing among adolescents. This is problematic since consuming nicotine at an early age can cause harmful effects in developing teenager's brain and health. Additionally, the use of e-cigarette has a possibility of leading to the use of cigarettes, which is more severe. There were many researches about e-cigarette and cigarette that mostly focused on finding and analyzing causes of smoking using conventional statistics. However, there is a lack of research on developing prediction models, which is more applicable to anti-smoking campaign, about e-cigarette and cigarette. In this paper, we research the prediction models that can be used to predict an individual e-cigarette user's (including non-e-cigarette users) intention to smoke cigarettes, so that one can be early informed about the risk of going down the path of smoking cigarettes. To construct the prediction models, five machine learning (ML) algorithms are exploited and tested for their accuracy in predicting the intention to smoke cigarettes among never smokers using data from the 2018 National Youth Tobacco Survey (NYTS). In our investigation, the Gradient Boosting Classifier, one of the prediction models, shows the highest accuracy out of all the other models. Also, with the best prediction model, we made a public website that enables users to input information to predict their intentions of smoking cigarettes.

Class probabilities predicted by most multiclass classifiers are uncalibrated, often tending towards over-confidence. With neural networks, calibration can be improved by temperature scaling, a method to learn a single corrective multiplicative factor for inputs to the last softmax layer. On non-neural models the existing methods apply binary calibration in a pairwise or one-vs-rest fashion. We propose a natively multiclass calibration method applicable to classifiers from any model class, derived from Dirichlet distributions and generalising the beta calibration method from binary classification. It is easily implemented with neural nets since it is equivalent to log-transforming the uncalibrated probabilities, followed by one linear layer and softmax. Experiments demonstrate improved probabilistic predictions according to multiple measures (confidence-ECE, classwise-ECE, log-loss, Brier score) across a wide range of datasets and classifiers. Parameters of the learned Dirichlet calibration map provide insights to the biases in the uncalibrated model.

Prominently used in support vector machines and logistic regressions, kernel functions (kernels) can implicitly map data points into high dimensional spaces and make it easier to learn complex decision boundaries. In this work, by replacing the inner product function in the softmax layer, we explore the use of kernels for contextual word classification. In order to compare the individual kernels, experiments are conducted on standard language modeling and machine translation tasks. We observe a wide range of performances across different kernel settings. Extending the results, we look at the gradient properties, investigate various mixture strategies and examine the disambiguation abilities.

The EM algorithm is one of the most popular algorithm for inference in latent data models. The original formulation of the EM algorithm does not scale to large data set, because the whole data set is required at each iteration of the algorithm. To alleviate this problem, Neal and Hinton have proposed an incremental version of the EM (iEM) in which at each iteration the conditional expectation of the latent data (E-step) is updated only for a mini-batch of observations. Another approach has been proposed by Capp\'e and Moulines in which the E-step is replaced by a stochastic approximation step, closely related to stochastic gradient. In this paper, we analyze incremental and stochastic version of the EM algorithm as well as the variance reduced-version of Chen et. al. in a common unifying framework. We also introduce a new version incremental version, inspired by the SAGA algorithm by Defazio et. al. We establish non-asymptotic convergence bounds for global convergence. Numerical applications are presented in this article to illustrate our findings.

We consider the problem of training machine learning models in a risk-averse manner. In particular, we propose an adaptive sampling algorithm for stochastically optimizing the Conditional Value-at-Risk (CVaR) of a loss distribution. We use a distributionally robust formulation of the CVaR to phrase the problem as a zero-sum game between two players. Our approach solves the game using an efficient no-regret algorithm for each player. Critically, we can apply these algorithms to large-scale settings because the implementation relies on sampling from Determinantal Point Processes. Finally, we empirically demonstrate its effectiveness on large-scale convex and non-convex learning tasks.

Overlapping clusters are common in models of many practical data-segmentation applications. Suppose we are given $n$ elements to be clustered into $k$ possibly overlapping clusters, and an oracle that can interactively answer queries of the form "do elements $u$ and $v$ belong to the same cluster?" The goal is to recover the clusters with minimum number of such queries. This problem has been of recent interest for the case of disjoint clusters. In this paper, we look at the more practical scenario of overlapping clusters, and provide upper bounds (with algorithms) on the sufficient number of queries. We provide algorithmic results under both arbitrary (worst-case) and statistical modeling assumptions. Our algorithms are parameter free, efficient, and work in the presence of random noise. We also derive information-theoretic lower bounds on the number of queries needed, proving that our algorithms are order optimal. Finally, we test our algorithms over both synthetic and real-world data, showing their practicality and effectiveness.

We develop a progressive training approach for neural networks which adaptively grows the network structure by splitting existing neurons to multiple off-springs. By leveraging a functional steepest descent idea, we derive a simple criterion for deciding the best subset of neurons to split and a splitting gradient for optimally updating the off-springs. Theoretically, our splitting strategy is a second-order functional steepest descent for escaping saddle points in an $\infty$-Wasserstein metric space, on which the standard parametric gradient descent is a first-order steepest descent. Our method provides a new computationally efficient approach for optimizing neural network structures, especially for learning lightweight neural architectures in resource-constrained settings.

Logistic regression is a model for binary classification predictive modeling. The parameters of a logistic regression model can be estimated by the probabilistic framework called maximum likelihood estimation. Under this framework, a probability distribution for the target variable (class label) must be assumed and then a likelihood function defined that calculates the probability of observing the outcome given the input data and the model. This function can then be optimized to find the set of parameters that results in the largest sum likelihood over the training dataset. The maximum likelihood approach to fitting a logistic regression model both aids in better understanding the form of the logistic regression model and provides a template that can be used for fitting classification models more generally.

Ever wonder how we can apply machine learning algorithms to a problem in order to analyze, visualize, discover trends & find correlations within data? In this article, I'm going to discuss common steps for setting up a machine learning model as well as approaches in selecting the right model for your data. This article was inspired by common interview questions that were asked about how I go along with my approach with a data science problem and why I choose said model. Machine learning tasks can be classified into either supervised learning, unsupervised learning, semi-supervised learning & reinforcement learning. In this article we don't focus on the last two, however, I'll give some idea of what they're.

Probability is at the very core of a lot of data science algorithms. In fact, the solutions to so many data science problems are probabilistic in nature – hence I always advice focusing on learning statistics and probability before jumping into the algorithms. But I've seen a lot of aspiring data scientists shunning statistics, especially Bayesian statistics. It remains incomprehensible to a lot of analysts and data scientists. I'm sure a lot of you are nodding your head at this! Bayes' Theorem, a major aspect of Bayesian Statistics, was created by Thomas Bayes, a monk who lived during the eighteenth century. The very fact that we're still learning about it shows how influential his work has been across centuries!