Event Our offer of discount early-bird tickets for Minds Mastering Machines ends next Monday, so act now if you want to join us to learn how real organisations can exploit machine learning and artificial intelligence and save big. We'll be bringing together a fantastic lineup of experts and practitioners at our conference on September 30 and October 1, headlined by Facebook AI's London research manager Sebastian Riedel and machine-learning veteran Dr Lorien Pratt. And if you want to get deep, and save even more, you can also get early bird prices on our October 2 workshops, which cover: developing and deploying Neural Nets; text mining; developing with TensorFlow 2; and getting machine learning into production using containers and devops. The venue is the palatial QE II Conference Center, in London, England, and the event runs from September 30 to October 2. As usual there will be excellent food right the way through, as well as our first-day drinks party, meaning you can connect with the speakers and your fellow attendees But remember, early bird prices expire next week, so to lock in your spot, head to the MCubed website now.

NASA introduced the term "digital twin" in a 2010 technology roadmap describing the tools of space travel. Almost a decade later, "digital twin" has emerged as a key tool. It enables a terrestrial space shot. We are being shown the global Industry 4.0's evolution of Digital Twin: from automation to autonomy. We talk about "digital transformation" in the Fourth Industrial Revolution, but "digital" has been around since the Third Revolution.

Let's start with a basic question, when you are in a hurry and want to know how something is done which of the two do you tend to do: refer a book about the topic or simply google your problem... If you just said "refer a book", we both know you're joking. This year has been all about artificial intelligence. We believe it's one thing everyone talks about but only a few companies have spoken it into existence. And those companies that have spoken it into existence are, needless to say, extremely happy with the results.

Intelligent, learning, autonomous machines are about to change the way we do business forever. But in a world where corporations or even executives may be liable in a civil or even criminal court for their decisions, who is responsible for decisions made by artificial intelligence (AI)? In the United States, courts are already having to wrestle with this science fiction scenario after an Arizona woman was killed by an experimental autonomous Uber vehicle. The European Commission recently shared ethical guidelines, requiring AI to be transparent, have human oversight and be subject to privacy and data protection rules. This sounds really good, but how will any of this be applied in practical situations?

CEOs and CFOs are decidedly more nervous when fielding questions about China during earnings calls this year. What's more, they are more likely to be deceptive with their answers. "Deception associated with questions on China has skyrocketed this quarter, up about 50% from last quarter and more than double a year ago," according to a study by text analytics provider Amenity Analytics. Amenity Analytics is one of a handful of companies that are applying natural language processing (NLP), sentiment analysis and machine learning to the financial sector, evaluating earnings calls and other public meetings to unearth information of value to an investor. It is also rare technology that offers a clear path to ROI.

An ever-present threat to any given country's national security is that of cybersecurity. There are always hackers that want to use technology for malicious purposes, not to say the long list of adversaries that a country can pile up along the years. That's so as what it is at stake is millions of sensible data from citizens, companies, directories, senior officers and members of the government, state's information and more. Unfortunately, not all Governments take this peril as seriously as they should, and the efforts towards creating cyber-defense strategies – in most countries – lack budget, personnel and even real, field knowledge. Before this absence of real policies, Artificial Intelligence might be well seen as a good starting point where to build the walls that keep out any possible threats. Governments and countries held sensible data of millions of unaware citizens, though their cyber-defense strategies leave much to be desired.

A deep-learning algorithm has been developed which can solve the Rubik's cube faster than any human can. It never fails to complete the puzzle, with a 100 per cent success rate and managing it in around 20 moves. Humans can beat the AI's mark of 18 seconds, the world record is around four seconds, but it is far more inefficient and people often require around 50 moves. It was created by University of California Irvine and can be tried out here. Given an unsolved cube, the machine must decide whether a specific move is an improvement on the existing configuration.

An AI has been studying the cookbooks and has taught itself how to make intriguing new pie recipes -- including Scotch egg pies and one with a salad filling. Working with a Sussex-based pie makers, the algorithm has produced thousands of recipes, five of which have been selected for production and will be going on sale. The AI works by looking for patterns in existing recipes and then trying to make its own based on what it learnt. While some of the early recipes it proposed were perhaps less-than-mouth-watering, with a little guidance it soon got the hang of cooking up new pie concepts. The experiment illustrates how artificial intelligence can provide new insights for small businesses and help dream up novel products to take to market.

A store themed around the work of "Astro Boy" manga artist Osamu Tezuka opened earlier this month in Tokyo's Asakusa district, putting an array of available products on display, from traditional Japanese crafts to artificial intelligence robots. The Tezuka Osamu Shop & Cafe is currently the only store, apart from the artist's memorial museum in western Hyogo Prefecture where he grew up, that sells character goods featuring his manga and anime, according to the shop's operator. With theme songs from his animation work playing in the background, the first floor displays approximately 300 types of merchandise, including wooden kokeshi (Japanese dolls) in the shape of characters including Astro Boy and his father figure Professor Ochanomizu, as well as ties featuring another masterpiece, "Phoenix," made in traditional Nishijin textiles. "Astro Boy" tells the stories of the adventures of a boy android with human emotions. The sci-fi manga series, serialized from 1952 to 1968 and also adapted into an animation series, has many fans in Asia and beyond.

Automated machine learning (AutoML) systems are helpful data science assistants designed to scan data for novel features, select appropriate supervised learning models and optimize their parameters. For this purpose, Tree-based Pipeline Optimization Tool (TPOT) was developed using strongly typed genetic programing (GP) to recommend an optimized analysis pipeline for the data scientist's prediction problem. However, like other AutoML systems, TPOT may reach computational resource limits when working on big data such as whole-genome expression data. We introduce two new features implemented in TPOT that helps increase the system's scalability: Feature Set Selector (FSS) and Template. FSS provides the option to specify subsets of the features as separate datasets, assuming the signals come from one or more of these specific data subsets. FSS increases TPOT's efficiency in application on big data by slicing the entire dataset into smaller sets of features and allowing GP to select the best subset in the final pipeline. Template enforces type constraints with strongly typed GP and enables the incorporation of FSS at the beginning of each pipeline. Consequently, FSS and Template help reduce TPOT computation time and may provide more interpretable results. Our simulations show TPOT-FSS significantly outperforms a tuned XGBoost model and standard TPOT implementation.