A team of scientists from the National Institute of Informatics in Japan have given up on making their AI smart enough to get into the University of Tokyo. The Todai Project -- Tokyo University's local nickname is "Todai" -- began in 2011, so we're sure at least some of them were disappointed by the decision. But, hey, this means robots aren't as smart as humans yet, and we don't have to fear an uprising in the near future. The researchers were hoping that the AI would score much higher than the 511 out of 950 it got last year when it took a standardized entrance exam in the country. Unfortunately, its overall results were pretty much the same for 2016.

Data science and machine learning have long been interests of mine, but now that I'm working on Fuzzy.io I need to keep on top of all the news in both fields. My preferred way to do this is through listening to podcasts. I've listened to a bunch of machine learning and data science podcasts in the last few months, so I thought I'd share my favorites: Every other week, they release a 10–15 minute episode where hosts, Kyle and Linda Polich give a short primer on topics like k-means clustering, natural language processing and decision tree learning, often using analogies related to their pet parrot, Yoshi. This is the only place where you'll learn about k-means clustering via placement of parrot droppings.

Something is happening in Bhaumau - a rural village in India's populous state of Uttar Pradesh where parents spend most of their time working in agriculture and as day labourers. Children, with no guidance from adults, are forming learning groups and with nothing more than a tablet computer preloaded with educational videos, stories and games, they are learning English and conducting science experiments. In the first three months of playing with the tablets there has been, according to the project's monitoring data, an 11% increase in pupils' core academic skills such as reading in children's mother tongue, reading and speaking in English, and science. Maybe even more important, children are figuring out how to navigate the digital world to find out answers to their questions and are more confident about speaking up. This is a radically different approach to using technology to advance learning.

We examine the effect of the Group Lasso (gLasso) regularizer in selecting the salient nodes of Deep Neural Network (DNN) hidden layers by applying a DNN-HMM hybrid speech recognizer to TED Talks speech data. We test two types of gLasso regularization, one for outgoing weight vectors and another for incoming weight vectors, as well as two sizes of DNNs: 2048 hidden layer nodes and 4096 nodes. Furthermore, we compare gLasso and L2 regularizers. Our experiment results demonstrate that our DNN training, in which the gLasso regularizer was embedded, successfully selected the hidden layer nodes that are necessary and sufficient for achieving high classification power.

The Obama administration may be headed for the exits, but it continues to focus on the impact of artificial intelligence on the economy and the nation at large. Artificial intelligence (AI) is an umbrella term for a group of technologies--including machine learning--that enable computers to learn new skills and capabilities based on the data they are exposed to, among other factors. The just-released report, titled "Preparing for the Future of Artificial Intelligence," does not downplay potential job lossdue to the advent of ever-smarter computers, but still posits that the technology will open up new career opportunities for those versed in it or who have higher-level skills. It also holds that public policy, especially re-training programs, can mitigate the negative impact of increasing automation by preparing displaced workers for other jobs. Prepared for the White House by the National Science and Technology Council's Subcommittee on Machine Learning and Artificial Intelligence, the report reiterates the conventional wisdom that AI-fueled automation will take over more jobs that could not be automated in the past.

Some videos are available only by enrolling in a Coursera or EdX class. It is free to do so, but sometimes the classes are no longer in session so you have to wait a couple of months, so you have no access. I'm going to be adding more videos from public sources and replacing the online course videos over time. I like using university lectures.

A team of researchers has failed to develop a robot smart enough to get into Japan's top university. The team, with members from the National Institute of Informatics, said Monday it was abandoning the effort began in 2011 to make a robot able to score well enough on entrance exams to achieve admission into the University of Tokyo, or Todai as it is commonly known. The researchers had hoped to create by March 2022 an artificial intelligence program capable of making the grade. The AI program dubbed Todai Robot had steadily improved its academic performance, but the team found a limit in its ability to understand various exam questions. "AI is not good at answering a type of question that requires the ability to grasp meaning in a broad spectrum," said Noriko Arai, a professor at the National Institute of Informatics.

SAP SE (NYSE: SAP) today announced three new courses delivered on the openSAP platform to guide users through the transformative effects experienced as a result of using iOS technology, machine learning and SAP S/4HANA to implement everyday business processes. These courses come in addition to the recent release of "Upgrade of Systems Based on SAP NetWeaver – Advanced Topics," which investigates the latest tools and features essential for SAP software system upgrades and maintenance. The new courses on openSAP will cover how the partnership between SAP and Apple is optimizing the use of iOS technology for end-to-end business processes. They also provide users an in-depth look at how machine learning is giving rise to new intelligent applications. SAP Fiori for iOS – An Introduction: The recent partnership forged between SAP and Apple will enable developers to build quickly their own native apps for Apple iOS devices.

The next frontier for artificial intelligence may involve teaching computers to distinguish albacore tuna from its yellowfin cousin. The Nature Conservancy, an environmental non-profit, is working with several Pacific Island nations and a big tuna fishing company to more easily count and identify fish caught at sea using cutting edge technology. The goal is to use trendy artificial intelligence techniques like deep learning to help fishermen reduce the number of protected animals like sharks and turtles that are accidentally caught along with the tuna. The Nature Conservancy hopes that the program could prevent overfishing and help threatened and endangered sea life recover without putting fishermen out of work. "We have real optimism that data science community can help us differentiate a turtle from a tuna, and flag when a shark comes on board," said Mark Zimring, a project director for The Nature Conservancy.

Combining deep neural networks with structured logic rules is desirable to harness flexibility and reduce uninterpretability of the neural models. We propose a general framework capable of enhancing various types of neural networks (e.g., CNNs and RNNs) with declarative first-order logic rules. Specifically, we develop an iterative distillation method that transfers the structured information of logic rules into the weights of neural networks. We deploy the framework on a CNN for sentiment analysis, and an RNN for named entity recognition. With a few highly intuitive rules, we obtain substantial improvements and achieve state-of-the-art or comparable results to previous best-performing systems.