A vectorization is basically the art of getting rid of explicit for loops whenever possible. With the help of vectorization, operations are applied to whole arrays instead of individual elements. The rule of thumb to remember is to avoid using explicit loops in your code. Deep learning algorithms tend to shine when trained on large datasets, so it's important that your code runs quickly. Otherwise, your code might take a long time to get your result.

They promised us the robots wouldn't attack… In what seems like dialogue lifted straight from the pages of a post-apocalyptic science fiction novel, researchers from the University of Massachusetts Amherst and Stanford claim they've developed an algorithmic framework that guarantees AI won't misbehave. The framework uses'Seldonian' algorithms, named for the protagonist of Isaac Asimov's "Foundation" series, a continuation of the fictional universe where the author's "Laws of Robotics" first appeared. According to the team's research, the Seldonian architecture allows developers to define their own operating conditions in order to prevent systems from crossing certain thresholds while training or optimizing. In essence, this should allow developers to keep AI systems from harming or discriminating against humans. Deep learning systems power everything from facial recognition to stock market predictions.

One of the world's top human champions of Go has decided to retire from playing the strategy game professionally, citing AI as a reason he feels he can no longer compete. South Korean Go player Lee Sedol gained international notoriety in March 2016 when he took on Google Deepmind's artificial intelligence AlphaGo. The machine won four out of five matches against Sedol, proving that AI is advanced enough to beat humanity at one of its most complex abstract strategy games. Sedol did not hide his sense of failure after his losses. "I don't know how to start or what to say today, but I think I would have to express my apologies first," he said after the third match.

A master player of the Chinese strategy game Go has decided to retire, due to the rise of artificial intelligence that "cannot be defeated". Lee Se-dol is the only human to ever beat the AlphaGo software developed by Google's sister company Deepmind. In 2016, he took part in a five-match showdown against AlphaGo, losing four times but beating the computer once. The South Korean said he had decided to retire after realising: "I'm not at the top even if I become the number one." "There is an entity that cannot be defeated," the 18-time world Go champion told South Korea's Yonhap news agency.

In 2018, Streamr announced a partnership with Daisy AI, Japan, an AI platform using blockchain for deep learning, based in Japan. Streamr intended to become Daisy AI's official data provider to exclusively sell data from Streamr's decentralized data Marketplace. Daisy AI planned to purchase data for a wide range of purposes, including forecasting stock and cryptocurrency prices, economy insights, footfall and traffic. Shohei Ohsawa, representative director of Daisy AI and associate professor at the University of Tokyo, recently made a series of racist and offensive statements on Twitter. He wrote that "Daisy does not hire Chinese people."

Today we can see how computer vision (CV) systems are revolutionizing whole industries and business functions with successful applications in healthcare, security, transportation, retail, banking, agriculture, and more. In 2019, we saw lots of novel architectures and approaches that further improved the perceptive and generative capacities of visual systems. To help you navigate through the overwhelming number of great computer vision papers presented this year, we've curated and summarized the top 10 CV research papers of 2019 that will help you understand the latest trends in this research area. The papers that we selected cover optimization of convolutional networks, unsupervised learning in computer vision, image generation and evaluation of machine-generated images, visual-language navigation, captioning changes between two images with natural language, and more. Subscribe to our AI Research mailing list at the bottom of this article to be alerted when we release new summaries. If you'd like to skip around, here are the papers we featured: Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available.

Besides "fake news", there's been another term which has caused many debates around it: deepfake. These seemingly realistic videos that are, in fact, manipulated have become more problematic lately, casting a shadow on the trust people have in media. Can blockchain and artificial intelligence (AI) be used to combat deepfake, to restore the public confidence back into the system? What is "deepfake", a term which combines'deep learning' and'fake'? According to Wikipedia, it's a technique for human image synthesis based on AI.

Retail health clinics have been part of the trend in making healthcare more convenient, and now another option is being offered -- testing for diabetic retinopathy. However, an ophthalmologist won't make the diagnosis at the clinic; instead, it will be made by an artificial intelligence (AI) system called IDx-DR. Testing will be offered through CarePortMD, the first retail health clinic to adopt this type of AI diagnostic technology, and offered at clinics inside Albertsons grocery stores. The second largest grocery chain in the United States, Albertsons added five CarePortMD clinics to stores in Delaware and Pennsylvania this past year. "Ours is a hybrid model of telehealth plus the convenience and access of a retail clinic, with the scalability and opportunity to coordinate with telemedicine." said Ashok Subramanian, MD, the CEO of CarePortMD.

Deep learning (DL) is hot. And it has cool tools to play with. Although scores of DL practitioners started their journey with TensorFlow, PyTorch has become an equally popular deep learning framework, since it was introduced by Facebook AI Research (FAIR) team, back in early 2017. Since its introduction, it has caught the attention of AI researchers and practitioners around the world and has matured significantly. In essence, PyTorch provides tremendous flexibility to a programmer about how to create, combine, and process tensors as they flow through a network (called computational graph) paired with a relatively high-level, object-oriented API. Raw TensorFlow, of course, provides a similar level of low-level flexibility but it is often difficult to master and troubleshoot.

On a sturdy workbench in seismologist Chris Marone's lab on the fifth floor of the geosciences building at Pennsylvania State University (Penn State) sits a large steel-framed machine with thick hydraulic pistons that force metal blocks and plates to grind past each other under extreme pressure. When the device is running, Marone sometimes closes the door to the lab so the loud bangs of "laboratory earthquakes" do not disrupt people across the hall. Lately, however, it has been the quieter sounds emanating from the machine that have caused a disruption in the field of seismology. In a recent spate of studies, researchers applied machine learning to acoustic emission data from Marone's earthquake machine, as well as from natural faults. The work led to the discovery of a new relationship between a fault's acoustic emissions and its physical characteristics, including its frictional state, its displacement rate, and the timing and magnitude of its next failure.