TensorFlow is an open source machine learning library developed at Google. TensorFlow uses data flow graphs for numerical computations. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them. In this post we will learn very basics of TensorFlow and we will build a Logistic Regression model using TensorFlow. The lowest level API - TensorFlow Core, provides you with complete programming control.

Data Science is one of the most-sought after professions today. Universities across the world are offering courses in this discipline which stands testimony to this emerging profession. There are a very few professionals with the required skill and the demand for data scientists is racing ahead. The tutorial wil give a brief understanding about Data Science. 'Implementing Machine Learning Algorithms on Larger Data Sets with Apache Mahout' have been widely covered in our course'Data Science'.

The characteristics (or numerical patterns) of a feature vector in the transform domain of a perturbation model differ significantly from those of its corresponding feature vector in the input domain. These differences - caused by the perturbation techniques used for the transformation of feature patterns - degrade the performance of machine learning techniques in the transform domain. In this paper, we proposed a nonlinear parametric perturbation model that transforms the input feature patterns to a set of elliptical patterns, and studied the performance degradation issues associated with random forest classification technique using both the input and transform domain features. Compared with the linear transformation such as Principal Component Analysis (PCA), the proposed method requires less statistical assumptions and is highly suitable for the applications such as data privacy and security due to the difficulty of inverting the elliptical patterns from the transform domain to the input domain. In addition, we adopted a flexible block-wise dimensionality reduction step in the proposed method to accommodate the possible high-dimensional data in modern applications. We evaluated the empirical performance of the proposed method on a network intrusion data set and a biological data set, and compared the results with PCA in terms of classification performance and data privacy protection (measured by the blind source separation attack and signal interference ratio). Both results confirmed the superior performance of the proposed elliptical transformation. 1 1. INTRODUCTION Feature vectors carry useful numerical patterns that characterize the original domain (or a sub original domain - input domain) formed by the feature vectors themselves. Machine learning algorithms generally utilize these patterns to generate classifiers, that can help make decisions from data, by using supervised or unsupervised learning techniques (Suthaharan, 2015).

In this paper, the problem of one-bit compressed sensing (OBCS) is formulated as a problem in probably approximately correct (PAC) learning. It is shown that the Vapnik-Chervonenkis (VC-) dimension of the set of half-spaces in $\mathbb{R}^n$ generated by $k$-sparse vectors is bounded below by $k \lg (n/k)$ and above by $2k \lg (n/k)$, plus some round-off terms. By coupling this estimate with well-established results in PAC learning theory, we show that a consistent algorithm can recover a $k$-sparse vector with $O(k \lg (n/k))$ measurements, given only the signs of the measurement vector. This result holds for \textit{all} probability measures on $\mathbb{R}^n$. It is further shown that random sign-flipping errors result only in an increase in the constant in the $O(k \lg (n/k))$ estimate. Because constructing a consistent algorithm is not straight-forward, we present a heuristic based on the $\ell_1$-norm support vector machine, and illustrate that its computational performance is superior to a currently popular method.

This paper deals with the unsupervised domain adaptation problem, where one wants to estimate a prediction function $f$ in a given target domain without any labeled sample by exploiting the knowledge available from a source domain where labels are known. Our work makes the following assumption: there exists a non-linear transformation between the joint feature/label space distributions of the two domain $\mathcal{P}_s$ and $\mathcal{P}_t$. We propose a solution of this problem with optimal transport, that allows to recover an estimated target $\mathcal{P}^f_t=(X,f(X))$ by optimizing simultaneously the optimal coupling and $f$. We show that our method corresponds to the minimization of a bound on the target error, and provide an efficient algorithmic solution, for which convergence is proved. The versatility of our approach, both in terms of class of hypothesis or loss functions is demonstrated with real world classification and regression problems, for which we reach or surpass state-of-the-art results.

We consider the problem of distributed statistical machine learning in adversarial settings, where some unknown and time-varying subset of working machines may be compromised and behave arbitrarily to prevent an accurate model from being learned. This setting captures the potential adversarial attacks faced by Federated Learning -- a modern machine learning paradigm that is proposed by Google researchers and has been intensively studied for ensuring user privacy. Formally, we focus on a distributed system consisting of a parameter server and $m$ working machines. Each working machine keeps $N/m$ data samples, where $N$ is the total number of samples. The goal is to collectively learn the underlying true model parameter of dimension $d$. In classical batch gradient descent methods, the gradients reported to the server by the working machines are aggregated via simple averaging, which is vulnerable to a single Byzantine failure. In this paper, we propose a Byzantine gradient descent method based on the geometric median of means of the gradients. We show that our method can tolerate $q \le (m-1)/2$ Byzantine failures, and the parameter estimate converges in $O(\log N)$ rounds with an estimation error of $\sqrt{d(2q+1)/N}$, hence approaching the optimal error rate $\sqrt{d/N}$ in the centralized and failure-free setting. The total computational complexity of our algorithm is of $O((Nd/m) \log N)$ at each working machine and $O(md + kd \log^3 N)$ at the central server, and the total communication cost is of $O(m d \log N)$. We further provide an application of our general results to the linear regression problem. A key challenge arises in the above problem is that Byzantine failures create arbitrary and unspecified dependency among the iterations and the aggregated gradients. We prove that the aggregated gradient converges uniformly to the true gradient function.

Although I am an experienced Data Scientist who knows well Python's stack for Data Science (scikit-learn, pandas, statsmodels, numpy, scipy, matplotlib, IPython), this book captured my attention and I have read a half of it during the first two days after getting the book. This book is easy to read for novices and experts alike (it does not contain a lot of math and wherever there are formulas they are not difficult to grasp), though some familiarity with Python packages comprising the Data Science stack will greatly facilitate material understanding. The writing style authors chose is excellent as it teaches readers in a very logical and pedagogically appealing way: the way of data pre-processing and analysis occur in projects that data scientists and engineers often encounter when aiming to solve the real-worlds tasks. The books begins with a description of how to install Python and various packages needed to run the code. The purpose of these packages is also explained.

Data analysis echo system has grown all the way from SQL's to NoSQL and from Excel analysis to Visualization. Today, we are in scarceness of the resources to process ALL (You better understand what i mean by ALL) kind of data that is coming to enterprise. Data goes through profiling, formatting, munging or cleansing, pruning, transformation steps to analytics and predictive modeling. Interestingly, there is no one tool proved to be an effective solution to run all these operations { Don't forget the cost factor here:) }. Things become challenging when we mature from aggregated/summarized analysis to Data mining, mathematical modeling, statistical modeling and predictive modeling.

If you enjoy reading, but still haven't foundyour next book to cozy up with, your smartphone might be able to suggest one. Artificial intelligence (AI) is now able to rank literature to predict the next bestseller – a kind of recommendation system, not based on metadata, but on the patterns and themes found in books. Publishers around the globe are mining all kinds of data, including what's in the books themselves, in search of the magic formula for evaluating a book's market potential. With more informed marketing, publishers hope to better target their customers. So, how does AI determine what we want to read?

Identifying arbitrary topologies of power networks in real time is a computationally hard problem due to the number of hypotheses that grows exponentially with the network size. A new "Learning-to-Infer" variational inference method is developed for efficient inference of every line status in the network. Optimizing the variational model is transformed to and solved as a discriminative learning problem based on Monte Carlo samples generated with power flow simulations. A major advantage of the developed Learning-to-Infer method is that the labeled data used for training can be generated in an arbitrarily large amount fast and at very little cost. As a result, the power of offline training is fully exploited to learn very complex classifiers for effective real-time topology identification. The proposed methods are evaluated in the IEEE 30, 118 and 300 bus systems. Excellent performance in identifying arbitrary power network topologies in real time is achieved even with relatively simple variational models and a reasonably small amount of data.