We study the problem of combining multiple bandit algorithms (that is, online learning algorithms with partial feedback) with the goal of creating a master algorithm that performs almost as well as the best base algorithm if it were to be run on its own. The main challenge is that when run with a master, base algorithms unavoidably receive much less feedback and it is thus critical that the master not starve a base algorithm that might perform uncompetitively initially but would eventually outperform others if given enough feedback. We address this difficulty by devising a version of Online Mirror Descent with a special mirror map together with a sophisticated learning rate scheme. We show that this approach manages to achieve a more delicate balance between exploiting and exploring base algorithms than previous works yielding superior regret bounds. Our results are applicable to many settings, such as multi-armed bandits, contextual bandits, and convex bandits. As examples, we present two main applications. The first is to create an algorithm that enjoys worst-case robustness while at the same time performing much better when the environment is relatively easy. The second is to create an algorithm that works simultaneously under different assumptions of the environment, such as different priors or different loss structures.

In this paper, a novel architecture for a deep recurrent neural network, residual LSTM is introduced. A plain LSTM has an internal memory cell that can learn long term dependencies of sequential data. It also provides a temporal shortcut path to avoid vanishing or exploding gradients in the temporal domain. The residual LSTM provides an additional spatial shortcut path from lower layers for efficient training of deep networks with multiple LSTM layers. Compared with the previous work, highway LSTM, residual LSTM separates a spatial shortcut path with temporal one by using output layers, which can help to avoid a conflict between spatial and temporal-domain gradient flows. Furthermore, residual LSTM reuses the output projection matrix and the output gate of LSTM to control the spatial information flow instead of additional gate networks, which effectively reduces more than 10% of network parameters. An experiment for distant speech recognition on the AMI SDM corpus shows that 10-layer plain and highway LSTM networks presented 13.7% and 6.2% increase in WER over 3-layer baselines, respectively. On the contrary, 10-layer residual LSTM networks provided the lowest WER 41.0%, which corresponds to 3.3% and 2.8% WER reduction over plain and highway LSTM networks, respectively.

Voice assistants are the latest focus among tech companies. For Apple, its Siri voice assistant remains popular, but for the first time in a while, it's feeling the heat from the competition. In a SurveyMonkey Audience poll for USA Today, 33 percent of those responding said they use Apple's Siri the most regularly. The survey's other findings follow a rough split of general smartphone marketshare: 19 percent of respondents used Google Assistant, 6 percent named Amazon's Alexa as their primary assistant and a middling 4 percent named Microsoft's Cortana as their main assistant. USA Today noted part of Siri's popularity comes from the assistant's head start.

Last year I wrote an overview post which defines a number of key concepts related to data science -- including data science itself -- and attempts to explain how these pieces fit together into a so-called "data science puzzle." As a new year begins, and a previous year worth of advances, insights, and accomplishments get rolled into our collective professional outlook, I thought it would be prudent to revisit this puzzle, noting and incorporating any changes and updates which may contribute to rearranging the puzzle for the foreseeable future, and to provide some addition commentary where warranted. Big Data is still important to data science. Take your pick of metaphors, but any way you look at it, Big Data is the raw material that has continues to fuel the data science revolution. As relates to Big Data, I believe that justification of data-acquisition and -retention from a business point of view, expectations that Big Data projects start providing actual financial returns, and the challenges related to data privacy and security will become the big Big Data stories not only of 2017 but moving forward in general.

"This applies to any deep-learning architecture, and the technique scales sublinearly, which means that the larger the deep neural network to which this is applied, the more the savings in computations there will be," said lead researcher Anshumali Shrivastava, an assistant professor of computer science at Rice. The research will be presented in August at the KDD 2017 conference in Halifax, Nova Scotia. It addresses one of the biggest issues facing tech giants like Google, Facebook and Microsoft as they race to build, train and deploy massive deep-learning networks for a growing body of products as diverse as self-driving cars, language translators and intelligent replies to emails. Shrivastava and Rice graduate student Ryan Spring have shown that techniques from "hashing," a tried-and-true data-indexing method, can be adapted to dramatically reduce the computational overhead for deep learning. Hashing involves the use of smart hash functions that convert data into manageable small numbers called hashes.

Knowing how good a set of predictions is, allows you to make estimates about how good a given machine learning model of your problem, In this tutorial, you will discover how to implement four standard prediction evaluation metrics from scratch in Python. You must estimate the quality of a set of predictions when training a machine learning model. As such, performance metrics are a required building block in implementing machine learning algorithms from scratch. These steps will provide the foundations you need to handle evaluating predictions made by machine learning algorithms.

Using artificial intelligence to predict how long we'll live, is now a step closer to reality thanks to new research. In the first study of its kind, machine learning was used to predict which patients would die within five years, with 69 per cent accuracy, by analysing imaging of patients' chests. The results are comparable with traditional predictions made by doctors. While the findings from the University of Adelaide study may sound a little macabre, they could have important implications for the early diagnosis of serious illness, enabling doctors to intervene with potentially life-saving treatments. "Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual," says lead author Dr Luke Oakden-Rayner, from the University of Adelaide's School of Public Health.

There is no denying the fact that Artificial Intelligence is the breakthrough technology of recent times. The machines have come a long way from assisting humans in mechanical operations to performing smarter tasks using cognitive intelligence. Every day, we are coming across interesting applications of AI. The ability of Deep Learning algorithms to learn and predict efficiently has opened the doors of possibilities. Nowadays, AI is impacting many other areas as well. In this blog post, we will discuss some niche applications of AI.

It was a miniscule change in parameters – just 2mm – but that tiny deviation delivered to a real robot in a recent hacking experiment could result in a massive product recall or even a major defect in an aircraft design in a real manufacturing scenario. Researchers at Trend Micro and Italy-based Politecnico di Milano today detailed the proof-of-concept attack they conducted on an ABB Robotics IRB140 industrial robot, exploiting a remote code vulnerability they found in the robot's controller software. They fed the robot a phony configuration file that modified its parameters for drawing a straight line. Instead of a perfectly straight line, the robot drew a slightly skewed one, following the 2mm change in instructions. "The code was working as expected, but with the wrong configuration," says Mark Nunnikhoven, Trend Micro's vice president of cloud research.