This resource is part of a series on specific topics related to data science: regression, clustering, neural networks, deep learning, decision trees, ensembles, correlation, Python, R, Tensorflow, SVM, data reduction, feature selection, experimental design, cross-validation, model fitting, and many more. To keep receiving these articles, sign up on DSC. Enjoy the reading, or become one of our bloggers and start posting articles and tutorials on DSC.

Artificial Intelligence systems are becoming the new warriors in disease diagnosis and can even accurately predict when you are going to die. Scientists at the University Of Adelaide in Australia have developed an Artificial Intelligence system that can accurately predict a human's life expectancy. The system was developed through research that examined the CT scan of 48 patients. Looking at the scans, the deep learning algorithms gave a'diagnosis' of whether the patient would die within 5 years. The prediction has a 69% accuracy rate, a score'similar' to the accuracy of human doctors.

Your mother was right: you are special. While each of us is a perfect little snowflake in our own right, that doesn't necessarily mean we possess world-shaking skills. But back in the lab, data scientists are cranking out algorithms that exceed human capability on a regular basis. About a year ago, Facebook CEO Mark Zuckerberg predicted that artificial intelligence (AI) would generally surpass humans in core sensory capabilities (like seeing and hearing) in about five to 10 years. AI still can't "actually look at the photo and deeply understand what's in it or look at the videos and understand what's in it," he said at the time.

Artificial intelligence (AI) has already managed to get a few amazing achievements. At its I/O 2017 conference, Google's CEO Sundar Pichai announced a new project that can be seen as "AI inception": auto-machine learning, or AutoML. Essentially, AI systems can now create better AI systems. Deep learning techniques, that allow AI to keep learning, involve passing data through complicated layers of neural networks. AutoML is Google's way of automating the time-consuming process by allowing AI to create these layers.

Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Google has never understood enterprise,

In this paper, we consider regression problems with one-hidden-layer neural networks (1NNs). We distill some properties of activation functions that lead to $\mathit{local~strong~convexity}$ in the neighborhood of the ground-truth parameters for the 1NN squared-loss objective. Most popular nonlinear activation functions satisfy the distilled properties, including rectified linear units (ReLUs), leaky ReLUs, squared ReLUs and sigmoids. For activation functions that are also smooth, we show $\mathit{local~linear~convergence}$ guarantees of gradient descent under a resampling rule. For homogeneous activations, we show tensor methods are able to initialize the parameters to fall into the local strong convexity region. As a result, tensor initialization followed by gradient descent is guaranteed to recover the ground truth with sample complexity $ d \cdot \log(1/\epsilon) \cdot \mathrm{poly}(k,\lambda )$ and computational complexity $n\cdot d \cdot \mathrm{poly}(k,\lambda) $ for smooth homogeneous activations with high probability, where $d$ is the dimension of the input, $k$ ($k\leq d$) is the number of hidden nodes, $\lambda$ is a conditioning property of the ground-truth parameter matrix between the input layer and the hidden layer, $\epsilon$ is the targeted precision and $n$ is the number of samples. To the best of our knowledge, this is the first work that provides recovery guarantees for 1NNs with both sample complexity and computational complexity $\mathit{linear}$ in the input dimension and $\mathit{logarithmic}$ in the precision.

In retailer management, the Newsvendor problem has widely attracted attention as one of basic inventory models. In the traditional approach to solving this problem, it relies on the probability distribution of the demand. In theory, if the probability distribution is known, the problem can be considered as fully solved. However, in any real world scenario, it is almost impossible to even approximate or estimate a better probability distribution for the demand. In recent years, researchers start adopting machine learning approach to learn a demand prediction model by using other feature information. In this paper, we propose a supervised learning that optimizes the demand quantities for products based on feature information. We demonstrate that the original Newsvendor loss function as the training objective outperforms the recently suggested quadratic loss function. The new algorithm has been assessed on both the synthetic data and real-world data, demonstrating better performance.

Deep neural networks have been shown to be very successful at learning feature hierarchies in supervised learning tasks. Generative models, on the other hand, have benefited less from hierarchical models with multiple layers of latent variables. In this paper, we prove that hierarchical latent variable models do not take advantage of the hierarchical structure when trained with existing variational methods, and provide some limitations on the kind of features existing models can learn. Finally we propose an alternative architecture that do not suffer from these limitations. Our model is able to learn highly interpretable and disentangled hierarchical features on several natural image datasets with no task specific regularization or prior knowledge.