Has Deep Learning Made Traditional Machine Learning Irrelevant?

Autonomous machines shining lasers at ants -- and spraying water at bewildered cats -- for the amusement of cackling grandchildren. Hobbyists are just getting started with deep-learning technologies that give them cheap, off-the-shelf capabilities that put Ronald Reagan's Star Wars program to shame. In the latest edition of the AI Podcast, NVIDIA engineer Bob Bond and Make: Magazine Executive Editor Mike Senese explain to host Michael Copeland how they've taken the once esoteric technology of deep learning and put it to work on offbeat projects that can be tackled on budgets of a few hundred bucks. "One of the big things that's happening -- and it's happening in real time right now -- is the technology is finally hitting a point where we, as consumers, have access to this type of capability," Senese says. Bond, a veteran engineer, is no technical novice.

On Dec. 7, a bag of popcorn, along with an Amazon Fire TV stick, left a Cambridge warehouse in the U.K. and 13 minutes later, both were accepted by an Amazon customer, one of two who had agreed to be part of the test program. A video posted by Amazon shows a fully autonomous drone, no humans involved, taking off from the warehouse and flying over fields to deposit the package just outside the customer's home. Amazon, with drones, aims to make deliveries in 30 minutes or less. Packages must weigh five pounds or less and can be delivered only during the day and in clear weather. The Seattle-based company plans to expand the trial to hundreds of users.

Recently, sentiment analysis has received a lot of attention due to the interest in mining opinions of social media users. Sentiment analysis consists in determining the polarity of a given text, i.e., its degree of positiveness or negativeness. Traditionally, Sentiment Analysis algorithms have been tailored to a specific language given the complexity of having a number of lexical variations and errors introduced by the people generating content. In this contribution, our aim is to provide a simple to implement and easy to use multilingual framework, that can serve as a baseline for sentiment analysis contests, and as starting point to build new sentiment analysis systems. We compare our approach in eight different languages, three of them have important international contests, namely, SemEval (English), TASS (Spanish), and SENTIPOLC (Italian). Within the competitions our approach reaches from medium to high positions in the rankings; whereas in the remaining languages our approach outperforms the reported results.

We address the problem of semi-supervised learning in relational networks, networks in which nodes are entities and links are the relationships or interactions between them. Typically this problem is confounded with the problem of graph-based semi-supervised learning (GSSL), because both problems represent the data as a graph and predict the missing class labels of nodes. However, not all graphs are created equally. In GSSL a graph is constructed, often from independent data, based on similarity. As such, edges tend to connect instances with the same class label. Relational networks, however, can be more heterogeneous and edges do not always indicate similarity. For instance, instead of links being more likely to connect nodes with the same class label, they may occur more frequently between nodes with different class labels (link-heterogeneity). Or nodes with the same class label do not necessarily have the same type of connectivity across the whole network (class-heterogeneity), e.g. in a network of sexual interactions we may observe links between opposite genders in some parts of the graph and links between the same genders in others. Performing classification in networks with different types of heterogeneity is a hard problem that is made harder still when we do not know a-priori the type or level of heterogeneity. Here we present two scalable approaches for graph-based semi-supervised learning for the more general case of relational networks. We demonstrate these approaches on synthetic and real-world networks that display different link patterns within and between classes. Compared to state-of-the-art approaches, ours give better classification performance without prior knowledge of how classes interact. In particular, our two-step label propagation algorithm gives consistently good accuracy and runs on networks of over 1.6 million nodes and 30 million edges in around 12 seconds.

American technology giant Yahoo has said it believes hackers stole data from more than one billion accounts in August 2013 - in a breach separate from the one it previously disclosed affecting 500 million accounts. The company said the information stolen may include names, email addresses, phone numbers, birthdates and security questions and answers, but added bank account information and payment-card data were not affected. "As Yahoo previously disclosed in November, law enforcement provided the company with data files that a third party claimed was Yahoo user data. The company analysed this data with the assistance of outside forensic experts and found that it appears to be Yahoo user data. "Based on further analysis of this data by the forensic experts, Yahoo believes an unauthorised third party, in August 2013, stole data associated with more than one billion user accounts."

On Dec. 7, a bag of popcorn, along with an Amazon Fire TV stick, left a Cambridge warehouse in the U.K. and 13 minutes later, both were accepted by an Amazon customer, one of two who had agreed to be part of the test program. A video posted by Amazon shows a fully autonomous drone, no humans involved, taking off from the warehouse and flying over fields to deposit the package just outside the customer's home. Amazon, with drones, aims to make deliveries in 30 minutes or less. Packages must weigh 5 pounds or less and can only be delivered during the day and in clear weather. Amazon plans to expand the trial to hundreds of users.

A group of hospitals in Paris is exploring the use of data analytics and machine learning to predict patient admissions down to the hour. Four hospitals within the city's public hospital system are trialing a new data-driven approach that combs through 10 years of admissions data and deploys refined algorithms to predict admissions rates during certain times of the day, according to Forbes' contributed post. The project is expected to be rolled out to all 44 public hospitals in Paris, which meant data scientists had to create a scalable, open-sourced framework from scratch, while also considering the need for cloud-based storage to house additional data in the future. Although privacy laws in France presented some barriers, the model has helped hospitals adjust staffing levels during periods of heavy admissions. Clinical and administrative staff members use a simple, web-based to review predicted admission rates 15 days out.

Want to watch this again later? Report Need to report the video? Report Need to report the video? Need to report the video? This feature is not available right now.

Three years after CEO Jeff Bezos secured a massive dose of publicity with the announcement that Amazon was working on drone deliveries, that vision is a reality. First-ever #AmazonPrimeAir customer delivery is in the books. Check out the video: https://t.co/Xl8HiQMA1S Workers at a British fulfillment center are stuffing shoebox-sized packages--in one case with a Fire TV and some popcorn--and loading them into the belly of electric, quadcopter drones, which set off to the customer. The company claims the drones, guided by GPS and flying below 400 feet, can make deliveries within 30 minutes, from click to plop.