Last year, a strange self-driving car was released onto the quiet roads of Monmouth County, New Jersey. The experimental vehicle, developed by researchers at the chip maker Nvidia, didn't look different from other autonomous cars, but it was unlike anything demonstrated by Google, Tesla, or General Motors, and it showed the rising power of artificial intelligence. The car didn't follow a single instruction provided by an engineer or programmer. Instead, it relied entirely on an algorithm that had taught itself to drive by watching a human do it. Getting a car to drive this way was an impressive feat. But it's also a bit unsettling, since it isn't completely clear how the car makes its decisions.

Convolutional neural networks provide visual features that perform remarkably well in many computer vision applications. However, training these networks requires significant amounts of supervision. This paper introduces a generic framework to train deep networks, end-to-end, with no supervision. We propose to fix a set of target representations, called Noise As Targets (NAT), and to constrain the deep features to align to them. This domain agnostic approach avoids the standard unsupervised learning issues of trivial solutions and collapsing of features. Thanks to a stochastic batch reassignment strategy and a separable square loss function, it scales to millions of images. The proposed approach produces representations that perform on par with state-of-the-art unsupervised methods on ImageNet and Pascal VOC.

Inspired by Wired articles by Oren Etzioni and Joe Lonsdale titled "Deep Learning Isn't a Dangerous Magic Genie. It's Just Math" and "AI and Robots Will Take Our Jobs -- But Better Ones Will Emerge for Us," I wanted to share six quick -- and down-to-earth -- thoughts about artificial intelligence. In conclusion, the printing press, sewing machines, combustion engines, ATMs, and "gig economy" (ugh) all caused massive disruption that was eventually very good in the end. Machine learning will have a much bigger, broader impact, and it will take less time. It is inevitable, and it is already happening (though it's still early days). I give thanks to folks like Etzioni and Lonsdale who are encouraging healthy debates based on the real ground truth, brick by brick.

Last week, I spoke at the Swiss Mobile Association. The event was held at one of the oldest cross-functional research institutes Gottlieb Duttweiler Institute just outside Zurich. Prior to being involved in IoT and AI, I worked for many years in Telecoms. I believe that from an innovation standpoint – we are living in a post-mobile world. Today, just as the Web itself, Mobile is a mature industry.

Artificial intelligence is gaining ground in healthcare. In 2012, there were fewer than 20 artificial intelligence startups focused on healthcare; last year there were almost 70, according to CB Insights. Additionally, the AI for healthcare sector is expected to drive overall AI market growth over the next six years, according to a MarketsandMarkets report. The overall AI market is expected to grow at a compound annual growth rate of 62.9 percent from 2016 to 2022, when it's projected to reach $16.6 billion. Here, six health IT company executives discuss how AI will impact healthcare over the next 20 years.

The news seemed to arrive all at once, even though Canada has long been at the forefront of technology, from Vancouver's film studios to Montreal's world-class animation talent. But in the past few months, things were different. Everyone seemed to come together (a rarity) and as a group, all parties were thinking ahead (another rarity): academia, government, non-profit organizations and businesses all came out in strong support of artificial intelligence research and development. The announcements were made in relatively quick succession: A new federal budget would provide $125 million to improve Canada's competitive and strategic advantage in AI. The University of Toronto's Vector Institute would hire roughly 25 new faculty and research scientists devoted to the field of artificial intelligence.

Once ordered alphabetically, each word can be referenced by its index, i.e. a, cat, chased, climbed, dog, saw, the, tree}. For this example, the neural network will have eight input neurons and eight output neurons. Let us assume that we decide to use three neurons in the hidden layer. This means that Winput and Woutput will be 8 3 and 3 8 matrices, respectively. Before training begins, these matrices are initialized to small random values as is usual in neural network training.

Google has a number of research projects underway aimed at making computers smarter and technically versatile. One of those projects involves teaching machines how to draw. On April 11, Google researchers released a technical paper describing "sketch-rnn", a neural network that has been trained by using thousands of crude human-drawn images to construct basic drawings of its own. One of the goals of the paper is to show that machines can be taught to draw certain things, like the sketch of a house, a tree or a dog, in a manner similar to humans. "As humans, we do not understand the world as a grid of pixels, but rather develop abstract concepts to represent what we see," wrote two of the papers authors, David Ha and Douglas Eck, who are researchers with Google Brain, the company's deep learning research group.

The science of Machine Learning (ML) has been around since the 1970s, but low horsepower processors and limited data forced the progress of Machine Learning to slow down in the 1980s. Ever since Big Data has enabled the use of unlimited "variety, volume, and velocity" business data, Machine Learning resurfaced as a powerful game changer in the world of software algorithms. Google's acquisition of UK-based Deep Mind resurrected the struggling field of Deep Learning (DL) and renewed the self-training possibilities of machines. In Deep Learning, smart algorithms can aid computers to learn from one layer of data and apply that learning to the next layer without programming intervention. While Machine Learning encompasses the entire field of learning algorithms, Deep Learning involves specific types of learning models where the human programmer is not required to train computers.

Google has taken a brilliant and unexpected step toward building an AI with more humanlike intuition, developing a computer capable of beating even expert human players at the fiendishly complicated board game Go. The objective of Go, a game invented in China more than 2,500 years ago, is fairly simple: players must alternately place black and white "stones" on a grid of 19 horizontal and 19 vertical lines with the aim of surrounding the opponent's pieces, and avoiding having one's own pieces surrounded. Mastering Go, however, requires endless practice, as well as a finely tuned knack of recognizing subtle patterns in the arrangement of the pieces spread across the board. Google's team has shown that the skills needed to master Go are not so uniquely human after all. Their computer program, called AlphaGo, beat the European Go champion, Fan Hui, five games to zero.