Deep learning has been widely applied and brought breakthroughs in speech recognition, computer vision, and many other domains. The involved deep neural network architectures and computational issues have been well studied in machine learning. But there lacks a theoretical foundation for understanding the approximation or generalization ability of deep learning methods generated by the network architectures such as deep convolutional neural networks having convolutional structures. Here we show that a deep convolutional neural network (CNN) is universal, meaning that it can be used to approximate any continuous function to an arbitrary accuracy when the depth of the neural network is large enough. This answers an open question in learning theory. Our quantitative estimate, given tightly in terms of the number of free parameters to be computed, verifies the efficiency of deep CNNs in dealing with large dimensional data. Our study also demonstrates the role of convolutions in deep CNNs.

There's a scene in Black Panther where Shuri, Letiticia Wright's character, hops into a car seat and remotely drives a sleek Lexus Sedan through the streets of Busan, Korea. While not quite a driverless car experience, remote driving offers a tantalizing imagining of the future of transportation. These are, after all, heady times for the transportation industry. Google and Uber are testing self-driving cars at this very moment. Drone deliveries from Amazon will happen sooner rather later.

Researchers from Osaka University have managed the reconstruction of plant branch structures, focusing on points of interest like branch structures under leaves. This has been achieved using advanced image analysis together with artificial intelligence technology. This represents one of the first applications of this type of technology to botany. The stud is of particular importance to fruit-bearing trees. By gaining insights into the growth of branches and leaves of individual trees, those whose livelihoods are based on the cultivation of fruit trees can learn about new methods for managing trees. This can help with maximizing fruit growth and with helping to maintain and protect the trees.

Insurers are increasingly interested in ways to make better use of data in their daily business decisions. Indeed, almost half of the CIOs we have surveyed this year mentioned that they are currently working on analytics including AI, machine learning, etc. It is also interesting to observe that a minority of insurers (5%) is not working on this topic currently. Our recent Model Insurer program helped us better understand what insurers are doing around data, analytics and AI. The awards' winners in this category have leveraged analytics and AI to improve various facets of their business.

After spending numerous evenings and weekends learning and coding for more than a year, you finally did it! You've now completed your data science program, earned your shiny certificate...now what? Chances are you were looking to get a job in data when you signed up for the course. So let's face this, it is time to get a job! The only thing that's standing between you and success is that first data science job offer.

Machine learning and data science are revolutionising the information technology industry and the way innovations are impacting our lives. With so much going on around these areas, it is often difficult to assimilate ideas and concepts around them. Additionally, the growing number of tools are overwhelming the these areas. This article discusses a visual mindset towards machine learning, that allows us to gain the best from the subject. One of the key factors that ML is dependent on, is the availability of right data for a project.

When it comes to the world of "big data", whoever has the most data has an advantage. And whoever has the most and best data scientists with the best tools to crunch that data usefully, build on that advantage. That's a key reason CIOs now look at all the data their companies collect and enlist data scientists and new tools to try to wring competitive advantage from what they learn. Entire economic blocs and countries are also upping the investment ante as they seek to slake their thirst for data and the riches machine learning and other AI technologies can bring. For example, the European Union's Digital Market Commission in April said it would increase its artificial intelligence R&D spending to $1.8 billion.

India and China have launched their first joint projects in the fields of Artificial Intelligence (AI) and Big Data. An AI-focused IT corridor was on Saturday launched by both countries in the northeastern Chinese city of Dalian, while on Sunday, a big data-focused IT cooperation platform was opened in the southwestern city of Guiyang. The two projects, backed by the Chinese government and India's National Association for Software and Services Companies (NASSCOM), are aimed at boosting cooperation between Indian software companies and Chinese firms in high-tech manufacturing in big data and Internet of Things projects. India has sought greater market access for IT companies in China, and the hope is that the two new initiatives will further open the door. The idea is to "help in bringing together the IT requirements of Chinese companies, particularly in Guiyang, with Indian companies which have solutions to offer," Indian envoy to China Gautam Bambawale said on Sunday.

Roundup Here's a quick roundup to keep you updated on what's been happening in AI, beyond what we've already covered, for your long weekend. It includes news of Samsung and Qualcomm setting up new AI research teams, why human radiologists are still better than machines and support for Amazon's Keras-MXNet backend. Hold your horses AI radiologists People are quick to believe that machines will soon replace radiologists because they think computers are much better at spotting abnormalities like tumors or clots in medical scans. But results reported by Stanford University shows that radiologists still trump AI. A group of researchers built a large convolutional neural network (CNN) with 169 layers to predict the probability of an abnormality appearing in a particular scan from the MURA (musculoskeletal radiographs) dataset. It collects 40,561 scans of the elbow, finger, forearm, hand, humerus, shoulder, and wrist of 12,173 patients.

It was barely three decades ago when the whole world was living in the fear of a full-blown nuclear war that could destroy anything and everything. The worst was just waiting to happen as the tensions between the superpowers - Soviet Union and the United States - had reached its peak during what we call the Cold War. One of the key reasons that the Cold War passed without the world descending into the havoc of a nuclear-powered third world war because of nuclear deterrence in the backdrop of a situation termed as MAD or'mutually assured destruction'. In any scenario of a potential first strike, either from the Soviets or the Americans, either side was aware of the fact that the other could and would retaliate, thus preventing an attack in the first place. However, things will not be the same if the military decides to incorporate artificial intelligence (AI) into their strategic decision making.