The Fujitsu Ltd. RoBoPin communication robot at the Combined Exhibition of Advanced Technologies in Japan on Oct. 4, 2016. Every year there is a new hot topic in tech. The difference between now and the past is that everything is becoming interconnected at a faster rate. We are entering an extremely critical time in history where society will change dramatically – how we work, live and play. Science fiction is morphing into reality. Flying cars exist, cars that drive themselves are on the road, and artificial intelligence that automates our lives is here.

In recent years, the best-performing systems in artificial-intelligence research have come courtesy of neural networks, which look for patterns in training data that yield useful predictions or classifications. A neural net might, for instance, be trained to recognize certain objects in digital images or to infer the topics of texts. But neural nets are black boxes. After training, a network may be very good at classifying data, but even its creators will have no idea why. With visual data, it's sometimes possible to automate experiments that determine which visual features a neural net is responding to.

Google just made a major upgrade to its Translate app. The company is now using a new technology called neural machine translation -- which aims to make computer-generated translations more similar to those done by humans -- to power its translations in seven new languages. Google says the update should make translations in those languages much more accurate and easier to understand. The company previously rolled out this technology for Chinese to English translations in September. Now, Google is using the same technology to power translations to and from English in French, German, Spanish, Portuguese, Japanese, Korean and Turkish.

When an app claims to be powered by "artificial intelligence" it feels like you're in the future. What does that really mean, though? We're taking a look at what buzzwords like AI, machine learning and neural networks really mean and whether they actually help improve your apps. Just recently, Google and Microsoft both added neural network learning to their translation apps. Google said it's using machine learning to suggest playlists. Todoist says it's using AI to suggest when you should finish a task.

Robots are taking over your job…and there's nothing you can do. By Amro Zakaria Abdu Human advancement throughout history can largely be credited to our ability to invent machines that increase our productivity and efficiency. Those tools allowed us to overcome the physical limitations of the human body and that of the animals we used, and as a result, territories were conquered, societies reshaped, and the dream of economic prosperity became a reality for millions. At the turn of the 19th century, the U.S. was a nation of farmers--39 percent of the population earned their livelihood through farming. The tractor was then introduced, resulting in profound changes such as the total replacement of work animals, consolidation of farms as seen in the increase in the average farm size from 60 to 200 hectares by the 1940's. Furthermore, the percentage of the population working in farming dropped to under 2 percent by the end of the century.

Chatbots can help secure purchases but also ship packages. Just in time for the holidays, UPS has launched a beta version of its chatbot that will mimic human conversations to help users find shipping locations, learn shipping rates and track packages. Available through Facebook Messenger, Skype and Amazon, UPS' release said it is different from the UPS website or mobile app in that users can use brief phrases like "shipping rates" and receive a voice response. "We see chatbots becoming an important communication channel for our customers over the next few years, and we're setting the stage for the incorporation of artificial intelligence throughout our customer-facing technologies," Stuart Marcus, UPS vice president of customer technology marketing, said in a release. Investing in a chatbot is likely useful for UPS' functionality, especially at times of high volume, such as the holiday rush.

Tuesday, 22nd of November, 2016, Baltimore, MD - Insilico Medicine, Inc and YMK Photonics, Inc announced today a research collaboration and business cooperation to develop photonics quantum computing and accelerated deep learning techniques for drug discovery, biomarker development and aging research. On the 15th of November, 2016 in the presence of over 800 YMK employees, customers, partners and investors, the Chairman of YMK Holdings, Mr. Kim Young Mo, the CEO of Insilico Medicine, Alex Zhavoronkov, PhD and Head of Insilico Korea, Professor Youngsook Park signed a memorandum of understanding to pursue mutual benefit in deep learning and cognitive photonics computing. "YMK is pursuing a very big vision. Extending healthy human longevity is not only the most altruistic cause, but a pressing socio-economic necessity. Insilico Medicine made substantial advances in applying deep learning techniques to drug development and aging research, but to accelerate the process and simulate entire human bodies and populations or generate optimal molecular structures, they could really benefit from our expertise in quantum computing. This collaboration is the first step towards cognitive quantum photonic computation for drug discovery and longevity research", said Mr. Kim Young Mo, the Chairman of YMK Holdings.

As artificial intelligence advances, the goal for modern tech companies is to build AI software that thinks for itself without human intervention. Towards that end, Amazon Web Services just picked MXNet, as its favored deep-learning framework to facilitate that work, according to a blog post Tuesday by Amazon chief technology officer Werner Vogels. Deep learning, as detailed in Fortune earlier this year, is a subset of AI that involves the use of software known as neural networks. Within this realm, software learns by churning through vast reams of data with the help of algorithms--not human programmers--to sort it out. Vogels said AWS will provide software code, documentation, and invest in the development of MXnet and the ecosystem of companies supporting it.

The most valuable part of machine learning is predictive modeling. This is the development of models that are trained on historical data and make predictions on new data. This cheat sheet contains my best advice distilled from years of my own application and studying top machine learning practitioners and competition winners. With this guide, you will not only get unstuck and lift performance, you might even achieve world-class results on your prediction problems. Note, the structure of this guide is based on an early guide that you might fine useful on improving performance for deep learning titled: How To Improve Deep Learning Performance.

In many respects, it is. For sure, the recent successes of deep learning have been amazing: we went from being really terrible at supervised learning on perceptual problems (image classification, speech recognition) to being really good at it. Deep learning has been transformative for many subfields of machine learning. But here's the thing: lots of people, most of them not directly involved with deep learning research, tend to extrapolate too much from these recent successes. For instance, when we started achieving below 4% top-5 error on the ImageNet classification task, people started claiming that we had "solved" computer vision.