Adversarial training provides a means of regularizing supervised learning algorithms while virtual adversarial training is able to extend supervised learning algorithms to the semi-supervised setting. However, both methods require making small perturbations to numerous entries of the input vector, which is inappropriate for sparse high-dimensional inputs such as one-hot word representations. We extend adversarial and virtual adversarial training to the text domain by applying perturbations to the word embeddings in a recurrent neural network rather than to the original input itself. The proposed method achieves state of the art results on multiple benchmark semi-supervised and purely supervised tasks. We provide visualizations and analysis showing that the learned word embeddings have improved in quality and that while training, the model is less prone to overfitting.

These data are often available in large quantities across U.S. states and require Big Data techniques to uncover complex hidden patterns. We design two different classes of models suitable for large-scale analysis of mortality data, a Hadoop-based ensemble of random forests trained over N-grams, and the DeepDeath, a deep classifier based on the recurrent neural network (RNN). We apply both classes to the mortality data provided by the National Center for Health Statistics and show that while both perform significantly better than the random classifier, the deep model that utilizes long short-term memory networks (LSTMs), surpasses the N-gram based models and is capable of learning the temporal aspect of the data without a need for building ad-hoc, expert-driven features. Many of the scientific discussions and studies in biomedical and healthcare domains address tasks whose end goal is to prevent death or diseases. Since the emergence of the big data science, numerous machine learning based techniques and technologies have been proposed and applied to improve human health by solving different computational challenges that we face today.

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

Given the ubiquity of fakes among re-sellers, buyers often examine pre-owned fashion to deduce authenticity, often analyzing the stitching, font size and interior labels. But sometimes, a copy is just so well-made that the human eye can't tell it from the original. Entrupy is a portable scanning device that instantly detects imitation designer bags by taking microscopic pictures that take into account details of the material, processing, workmanship, serial number, and wear/tear. It then employs the technique of deep learning to compare the images against a vast database that includes top luxury brands and if the bag is deemed authentic, users immediately get a Certificate of Authenticity. After launching as a paid service in September 2016, the New York-based venture now has over 130 paid customers, almost all of whom are American businesses drawn to the 97.1 percent accuracy rate, explained Entrupy CEO Vidyuth Srinivasan.

Unsupervised learning is the Holy Grail of Deep Learning. The goal of unsupervised learning is to create general systems that can be trained with little data. Today Deep Learning models are trained on large supervised datasets. Meaning that for each data, there is a corresponding label. In the case of the popular ImageNet dataset, there are 1M images labeled by humans.

Last week we discussed how doctors perform medicine, and what parts of the process are worth automating. It turns out that deep learning is a very good match for some of the most time consuming (and therefore costly) parts of medicine: the perceptual tasks. We also saw that many decisions simply fall out of the perceptual process; once you have identified what you are seeing or hearing, there is no more "thinking" work to do. In fact, the answers these systems arrive at can be superhuman. "In situations where the only information required to make the decision is in the signal itself, machine learning wins by a small margin." It turns out that quite a large subset of medical tasks are like this, which we will explore in more detail today. To begin with we should recognise that automating a subset of medical tasks is not the same as automating all of medicine.

Adam Geitgey writes about machine learning, deep learning, image and speech recognition, and related topics on his blog. He covers both theory and practice, and focuses on how developers are able to start using these technologies quickly. But the number one question I get asked is "How in the world do I get all these open source libraries installed and working on my computer?" To that end, Geitgey has put together a virtual machine image, which he outlines in this recent post. Based on Ubuntu Linux Desktop 16.04 LTS 64-bit, the image includes the following tools and libraries, among others: If you launch PyCharm Community Edition from the left sidebar, there are several pre-created projects you can open.

It is no secret that artificial intelligence (AI) and machine learning have advanced radically over the last decade, yet somewhere between better algorithms and faster processors lies the increasingly important task of engineering systems for maximum performance--and producing better results. The problem for now, says Nidhi Chappell, director of machine learning in the Datacenter Group at Intel, is that "AI experts spend far too much time preprocessing code and data, iterating on models and parameters, waiting for training to converge, and experimenting with deployment models. Each step along the way is either too labor-and/or compute-intensive." The research and development community--spearheaded by companies such as Nvidia, Microsoft, Baidu, Google, Facebook, Amazon, and Intel--is now taking direct aim at the challenge. Teams are experimenting, developing, and even implementing new chip designs, interconnects, and systems to boldly go where AI, deep learning, and machine learning have not gone before.

Researchers used scanning electron microscope images of nanometers-thick mouse brain slices to reconstruct cells into a neocortex structure (center), whose various cell types appear in different colors. For decades, scientists have worked toward the'holy grail' of finding a cure for cancer. While significant progress has been made, their efforts have often been worked on as individual entities. Now, as organizations of all kinds seek to put the massive amounts of data they take in to good use, so, too, are the health care industry and the U.S. federal government. The National Cancer Institute (NCI) and the U.S. Department of Energy (DOE) are collaborating on three pilot projects that involve using more intense high-performance computing at the exascale level, which is the push toward making a billion billion calculations per second (or 50 times faster than today's supercomputers), also known as exaFLOPS (a quintillion, 1018, floating point operations per second).

Our fourth installment of Research for Practice covers two of the hottest topics in computer science research and practice: cryptocurrencies and deep learning. First, Arvind Narayanan and Andrew Miller, co-authors of the increasingly popular open access Bitcoin textbook, provide an overview of ongoing research in cryptocurrencies. This is a topic with a long history in the academic literature that has recently come to prominence with the rise of Bitcoin, blockchains, and similar implementations of advanced, decentralized protocols. These developments--and colorful exploits such as the DAO vulnerability in June 2016--have captured the public imagination and the eye of the popular press. In the meantime, academics have been busy, delivering new results in maintaining anonymity, ensuring usability, detecting errors, and reasoning about decentralized markets, all through the lens of these modern cryptocurrency systems.