At the first "AI for Good Global Summit" in Geneva, the Chairman of the Board of Management of AUDI AG said in front of UN experts: "We need a discourse in society that looks at the enormous potential of piloted and autonomous driving in relation to the ethical and legal questions." Artificial intelligence (AI) is regarded by experts from science and industry as a key technology for piloted and autonomous driving. It assists the car in perceiving and interpreting its environment, and deriving decisions from this. But who is liable in the case of accidents, and how should the car driving autonomously behave in a situation of unavoidable danger? Prof. Stadler: "We take the concerns of the public seriously and are facing up to the challenges associated with this."

Artificial intelligence start-up Wyzerr announced today the launch of an online platform to create feedback surveys that feel like games. The surveys, which the company calls "Smart Forms," can capture up to 25 questions in under 60 seconds. Wyzerr's Smart Forms were developed based on playful design principals and rules of engagement. Wyzerr spent the past 2 years piloting their gammified surveys with large enterprises like Walmart, Unilever, and Volkswagen. The key learnings from those pilots served as the inspiration for the company's new online Smart Form builder, which guides customers in creating effective feedback campaigns.

The security of peoples' finances is as important as the protection of their personal data (87%) by their bank A lack of understanding and trust in technology is stalling mainstream adoption of innovative new services which could make millions of people's daily lives simpler and more secure, new HSBC research reveals. The HSBC research report, Trust in Technology, shows that technologies such as fingerprint recognition, voice recognition and robo-advice hold enormous potential for uses from bank security to mobile payments and investment advice, yet millions of people who could benefit do not trust them because they do not understand them well enough. HSBC has commissioned a study of more than 12,000 people in 11 countries and territories looking at their perceptions and use of technology. The study of more than 12,000 people in 11 countries reveals four in every five people (80%) believe that technology makes their lives easier but less than half (46%) trust fingerprint recognition to replace their password, despite it being recognised to be at least five1 times more secure and significantly more convenient than traditional passwords. Eighty-four per cent of respondents say they would share their personal data with their bank if it meant getting a better service, underlining the need to educate consumers about the specific benefits of new technologies.

Would you like to read a story about why Shopify is making its chatbot marketing assistant free to merchants? Type'yes,' send a thumbs up emoji, er, well actually, just scroll down to learn more. Techopia may not yet be run by bots (deep down, we're paranoid luddites) but Shopify believes it can reduce price barriers to growing a business by giving all of its merchants its first employee for free. Kit, which Shopify acquired last spring, is an automated assistant that can set up marketing campaigns, re-engage with customers via email, refresh social channels and run reports, all through Facebook Messenger or text message conversations with merchants. Before today, businesses on Shopify had to pay $10 per month for Kit's services, but the Ottawa e-commerce giant announced today that the firm would make its virtual assistant free to all merchants on its platform.

PALO ALTO, Calif., June 06, 2017 --Cloudera, Inc., (NYSE: CLDR), the leading provider of the modern platform for machine learning and advanced analytics built on the latest open source technologies, announced that Inova Translational Medicine Institute (ITMI), a global leading medical research institute, has deployed Cloudera Enterprise to securely analyze massive collections of clinical and genomic data at unprecedented speeds and scale for faster innovations in translational medicine research. As part of the Inova Center for Personalized Health (ICPH), ITMI's team of leading scientists, researchers, analysts and collaborators use machine learning algorithms on terabytes of clinical and genomic information to identify the genetic links to diseases. They make discoveries from the data insights and, in collaboration with the treating physician, develop personalized treatment plans for patients. This approach is also known as precision medicine and has the power to help patients live longer, healthier lives. Genetics plays a role in the majority of leading causes of death in the United States, including heart disease, cancer and diabetes.

Deep learning, a computing construct based loosely on the architecture of the human brain, has emerged as one of the most promising enabling technologies in the world of artificial intelligence (AI). Although many of the concepts underlying AI and technological biomimicry of human intelligence are over 50 years old, deep learning's growth today is the result of a rather sudden convergence of three key trends: big, even colossal, data generation; advancements in hardware capabilities; and improvements in algorithms. According to a new report from Tractica, deep learning has been commercialized into more than 100 distinct use cases to date, touching virtually every industry. Tractica forecasts that deep learning software revenue will grow from $655 million in 2016 to $34.9 million worldwide by 2025. "Businesses around the world are beginning to harness deep learning due to its ability to drive efficiencies in the form of speed, accuracy, agility, and access in several key areas," says principal analyst Jessica Groopman.

In late February, the UN-Secretary General held a press conference, highlighting the risk of starvation in East Africa and the necessity to raise funds to address the emergency situations in Somalia and South Sudan. Drought has been back in these countries and their neighbours since 2016, leading to a huge current food crisis. While governments are trying to handle the situation, how could technology innovations help prevent starvation and improve agriculture management in the future? We met with Laurent Rivière, a French 30 years-old entrepreneur, who shared with us his view on the subject with a combination of engineer pragmatism and changemaker idealism . Founder and CEO at Sunbirds for two years, he explained to us how his "bird of the sun," his solar drone, is addressing the agriculture challenges of the 21st century.

Ride service companies like Uber and Lyft are focused on the technology of self-driving cars, but what about everythingn else? Lyft is now exploring the passenger aspect via a partnership with self-driving software firm nuTonomy. The idea, the companies say in a news release, is to gain knowledge on areas like passenger comfort and safety, "from routing and booking to the performance of the driving system and how it interacts and communicates with the rider." Lyft and nuTonomy will be doing R&D in the Boston area at the Raymond L. Flynn marine park and nearby at Seaport and Fort Point. During trials, "an engineer from nuTonomy rides in each of its vehicles during testing to observe system performance and assume control if needed," the company said. The company is also running its own small self-driving trials on a small stretch of road in South Boston.

The global text analytics market size is estimated to grow from $3.97 billion in 2017 to $8.79 billion by 2022, at a Compound Annual Growth Rate (CAGR) of 17.2%. The customer experience management (CEM) is expected to hold the largest market share during the forecast period. Among the various applications in the text analytics market, the CEM application is expected to hold the largest market share during the forecast period. Text mining is the most traditional application in customer service and is frequently utilized to improve customer experience through various information sources. Today, text analytics is implemented to offer quick, computerized feedback to the clients, which significantly reduces dependency on executives for resolving issues.

Data preprocessing is a fundamental part of any machine learning project and often more time is spent on the data preparation than on the actual machine learning. While some preprocessing tasks are problem specific many others such as partitioning data into training and test folds, stratifying samples or building mini-batches are generic. The following Canonical Pipeline shows the processing steps common for deep-learning in vision. A Reader reads sample data stored in text files, Excel or Pandas tables. The Splitter then partitions data into training, validation and test folds and performs stratification if needed. Usually not all image data can be loaded into memory and a Loader loads images on demand. These images are often processed by a Transformer, for resizing, cropping or other adjustments. Furthermore, to increase the training set additional images are synthesized by randomly augmenting (flipping, rotating, …) images using an Augmenter. Efficient, GPU-based machine learning demands that image and label data are grouped in mini-batches via a Batcher before passed on to the Network for training or inference. Finally, to keep track of the training progress, usually a Logger is employed to write training losses or accuracies to a log file. Some machine learning frameworks such as Keras provide (some of) these preprocessing components hidden behind an API that considerably simplify network training if it fits the task at hand. See the following excerpt of a Keras example to train a model with augmentation. However, what if an image format, augmentation or other preprocessing capability is needed that is not provided by the API? Extending a library such as Keras or others is not trivial and the common approach is to (re)implement the required functionality – often in a quick-and-dirty fashion. But implementing a robust data pipeline that loads, transforms, augments and processes images on demand is challenging and time consuming. The following excerpt from a nuts-ml example shows a pipeline for network training, where the operator defines the flow of data. In the example above training images are augmented, pixel values re-ranged, and the samples shuffled before building batches for network training. Finally, the mean over the batch-wise training losses is computed and printed. The resulting code is more readable and can readily be modified to experiment with different preprocessing schemes. Task-specific functions can easily be implement as nuts and added to the data flow. Any machine learning library that accepts mini-batches of Numpy arrays for training or inference is compatible. For more information about nuts-ml see the Introduction and have a look at the Tutorial. Bio: Stefan Maetschke (PhD) is a research scientist at IBM Research Australia where he develops machine learning infrastructure and models for medical image analysis.