On 10 October 2017, the development of a Digital Agenda of the Federal Statistical Office of Germany (Destatis) has started (Statistisches Bundesamt 2018). One of many topics that were intensively discussed was Machine Learning. In a meeting at 13-15 November 2017, the office and department heads of Destatis evaluated and prioritised 59 measures of the Digital Agenda according to their benefits and costs. A "Proof of Concept Machine Learning" was given high priority and classified as one of four lighthouse projects of the Digital Agenda. The content specification was "Proof of Concept Machine Learning - Set up Proof of Concept for Machine Learning, e.g. in business statistics, to perform automatic categorization and improve analysis potential". The deadline for completion of the project was set for mid-2018.

Differential equations are used as numerical models to describe physical phenomena throughout the field of engineering and science, including heat and fluid flow, structural bending, and systems dynamics. Although there are many other techniques for finding approximate solutions to these equations, this paper looks to compare the application of the Theory of Connections (ToC) with one based on Support Vector Machines (SVM). The ToC method uses a constrained expression (an expression that always satisfies the differential equation constraints), which transforms the process of solving a differential equation into an unconstrained problem, and is ultimately solved via least-squares. In addition to individual analysis, the two methods are merged into a new methodology, called constrained SMVs (CSVM), by incorporating the SVM method into the ToC framework to solve unconstrained problems. Numerical tests are conducted on three sample problems: one first order linear ODEs, one first order non-linear ODE, and one second order linear ODE. Using the SVM method as a benchmark, a speed comparison is made for all the problems by timing the training period, and an accuracy comparison is made using the maximum error and mean-squared error on the training and test sets. In general, ToC is shown to be slightly faster (by an order of magnitude or less) and more accurate (by multiple orders of magnitude) over the SVM and CSVM approaches.

In this paper, we propose a scalable algorithm for spectral embedding. The latter is a standard tool for graph clustering. However, its computational bottleneck is the eigendecomposition of the graph Laplacian matrix, which prevents its application to large-scale graphs. Our contribution consists of reformulating spectral embedding so that it can be solved via stochastic optimization. The idea is to replace the orthogonality constraint with an orthogonalization matrix injected directly into the criterion. As the gradient can be computed through a Cholesky factorization, our reformulation allows us to develop an efficient algorithm based on mini-batch gradient descent. Experimental results, both on synthetic and real data, confirm the efficiency of the proposed method in term of execution speed with respect to similar existing techniques.

Classifiers used in the wild, in particular for safety-critical systems, should not only have good generalization properties but also should know when they don't know, in particular make low confidence predictions far away from the training data. We show that ReLU type neural networks which yield a piecewise linear classifier function fail in this regard as they produce almost always high confidence predictions far away from the training data. For bounded domains like images we propose a new robust optimization technique similar to adversarial training which enforces low confidence predictions far away from the training data. We show that this technique is surprisingly effective in reducing the confidence of predictions far away from the training data while maintaining high confidence predictions and similar test error on the original classification task compared to standard training.

Often survey analysis collects data to try to identify response patterns leading to groupings of respondents with different characteristics as revealed by answers provided to survey questions. Without additional background information on respondents, it is often very difficult (and many times impossible) to verify the accuracy of groupings resulting from the analysis. This paper examines one such situation in which high school students in low-income neighbourhood schools in Bolivia responded to a standard periodic institutional survey and responses were analysed to better understand respondents' socioeconomic contexts. In this case study, the question to be answered was "can we identify the most impoverished students based on a 22 questions standard survey alone?". With no known dependent variable and an inability to objectively capture the socioeconomic condition of the students being surveyed, the task of coming to a conclusive answer becomes unfeasible as there is no way to validate at least some portion of the students identified as most impoverished.

This is part two of this series, find part one here - How to build a data science project from scratch. After scraping or getting the data, there are many steps to accomplish before applying a machine learning model. You need to visualize each of the variables to see distributions, find the outliers, and understand why there are such outliers. What can you do with missing values in certain features? What would be the best way to convert categorical features into numerical ones?

Islamophobic hate speech on social media inflicts considerable harm on both targeted individuals and wider society, and also risks reputational damage for the host platforms. Accordingly, there is a pressing need for robust tools to detect and classify Islamophobic hate speech at scale. Previous research has largely approached the detection of Islamophobic hate speech on social media as a binary task. However, the varied nature of Islamophobia means that this is often inappropriate for both theoretically-informed social science and effectively monitoring social media. Drawing on in-depth conceptual work we build a multi-class classifier which distinguishes between non-Islamophobic, weak Islamophobic and strong Islamophobic content. Accuracy is 77.6% and balanced accuracy is 83%. We apply the classifier to a dataset of 109,488 tweets produced by far right Twitter accounts during 2017. Whilst most tweets are not Islamophobic, weak Islamophobia is considerably more prevalent (36,963 tweets) than strong (14,895 tweets). Our main input feature is a gloVe word embeddings model trained on a newly collected corpus of 140 million tweets. It outperforms a generic word embeddings model by 5.9 percentage points, demonstrating the importan4ce of context. Unexpectedly, we also find that a one-against-one multi class SVM outperforms a deep learning algorithm.

Abstract--Despite the success of deep neural networks (DNNs) in image classification tasks, the human-level performance relies on massive training data with high-quality manual annotations, which are expensive and time-consuming to collect. There exist many inexpensive data sources on the web, but they tend to contain inaccurate labels. Training on noisy labeled datasets causes performance degradation because DNNs can easily overfit to the label noise. To overcome this problem, we propose a noisetolerant trainingalgorithm, where a meta-learning update is performed prior to conventional gradient update. The proposed meta-learning method simulates actual training by generating synthetic noisy labels, and train the model such that after one gradient update using each set of synthetic noisy labels, the model does not overfit to the specific noise. We conduct extensive experiments on the noisy CIFAR-10 dataset and the Clothing1M dataset. The results demonstrate the advantageous performance of the proposed method compared to several state-of-the-art baselines. I. INTRODUCTION One of the key reasons why deep neural networks (DNNs) have been so successful in image classification is the collections ofmassive labeled datasets such as ImageNet [19] and COCO [14].

The Sinkhorn distance, a variant of the Wasserstein distance with entropic regularization, is an increasingly popular tool in machine learning and statistical inference. We give a simple, practical, parallelizable algorithm NYS-SINK, based on Nystr\"om approximation, for computing Sinkhorn distances on a massive scale. As we show in numerical experiments, our algorithm easily computes Sinkhorn distances on data sets hundreds of times larger than can be handled by state-of-the-art approaches. We also give provable guarantees establishing that the running time and memory requirements of our algorithm adapt to the intrinsic dimension of the underlying data.

Matrix Product States (MPS), also known as Tensor Train (TT) decomposition in mathematics, has been proposed originally for describing an (especially one-dimensional) quantum system, and recently has found applications in various applications such as compressing high-dimensional data, supervised kernel linear classifier, and unsupervised generative modeling. However, when applied to systems which are not defined on one-dimensional lattices, a serious drawback of the MPS is the exponential decay of the correlations, which limits its power in capturing long-range dependences among variables in the system. To alleviate this problem, we propose to introduce long-range interactions, which act as shortcuts, to MPS, resulting in a new model \textit{ Shortcut Matrix Product States} (SMPS). When chosen properly, the shortcuts can decrease significantly the correlation length of the MPS, while preserving the computational efficiency. We develop efficient training methods of SMPS for various tasks, establish some of their mathematical properties, and show how to find a good location to add shortcuts. Finally, using extensive numerical experiments we evaluate its performance in a variety of applications, including function fitting, partition function calculation of $2-$d Ising model, and unsupervised generative modeling of handwritten digits, to illustrate its advantages over vanilla matrix product states.