Sophie, the'digital human', greeted guests at the recent US launch of Air New Zealand's global marketing campaign A Better Way to Fly. Sophie showed her advanced emotional intelligence and responsiveness as she answered questions about New Zealand as a tourist destination and the airline's products and services. Air New Zealand worked with Soul Machines to create'Sophie' as it explores how artificial intelligence and other disruptive technologies can be used to help travelers. Soul Machines says the technology behind Sophie uses neural networks and brain models to bring its digital humans to life from their cloud based human computing engine which sits on top of an artificial Intelligence platform powered by IBM Watson. "We know there is a lot of pain points and frustrations that we can solve with digital innovation," says Air New Zealand CEO Christopher Luxon.

You can already use Google Home to order from Walmart, watch CBS All Access on your TV and hear daily news updates from former-VP Joe Biden. Now, Google has just updated Home with the ability to set reminders with your voice. The company confirmed to Engadget that (English) reminders would start rolling out Monday to the US, UK, Australia and Canada (French is coming soon). Google Home users can now set one-off and recurring reminders (daily or weekly) and even set reminders for months down the road. Google Assistant also understands context - if you ask it to remind you to walk the dog at 6 while you get ready for bed, for example, it will know to set the reminder to 6 AM.

In our high-vocational stress household, the most volcanic tension usually erupts over control of the PlayStation. I'm still – still – absorbed in the game of Fallout 4 I started a year ago, with thousands of hours spent on perfecting the aesthetics of post-apocalyptic settlement-building. My partner prefers a wordless immersion in the splattery worlds of first-person shooters and war games but we reconcile over rounds of two-player Diablo, fighting demons and hoarding treasure together. I've come a long way from the handheld Donkey Kong I cherished as a child, or the Pitfall caves I explored on a home PC, or the small parties of teens that gathered to play Sonic the Hedgehog on the loungeroom TV. The demands of fun are more complex now – but the need for fun remains the same.

Perhaps you've bumped into Mildred, Carla or Oscar on your recent travels. They're not real people but avatars of chatbots – concocted by Lufthansa, Avianca and Air New Zealand respectively – or artificial intelligence (AI) powered computer programs accessed on your smartphone that enable you to have a simulated conversation of sorts. Now airports are getting in on the act, and it's all part of a paradigm shift towards self-service and interactions with technologies that offer "personal" information to help us on our way through the terminal. It's a shift confirmed in the findings of the Passenger IT Trends Survey released by Sita, the provider of much of the digital infrastructure that underpins airport and airline communications and operations worldwide. The survey found that face-to-face check-in is now down to 46 per cent of passengers, and since last year's survey, self-service bag-tagging has risen from 31 per cent to 47 per cent. Almost a fifth of passengers now use self-service bag drop, and when it comes to ID control, 57 per cent of passengers would definitely use biometrics instead of a passport or boarding pass across the journey.

The International Joint Conference of Artificial Intelligence (or in short IJCAI) is the most established, important and leading scientific event in Artificial Intelligence. Established in 1969 as the first ever international conference on Artificial Intelligence (AI), it is an extension of the seminal (AI first) Dartmouth workshop in 1956 (for the interested, read the inspirational first papers on AI). Practically, much of the leading AI science and technology was presented during previous IJCAI conferences. Before we talk more about the upcoming IJCAI-ECAI-18 event in Stockholm, Sweden (July 13-19, 2018), let us share with you our first-hand experience of IJCAI 2017 in Melbourne, Australia. IJCAI 2017 brings together the brightest researcher and technologists from around the world.

SELF-DRIVING, personal assistants, cleaning robots, smart homes - these are just some examples of autonomous systems. With many such systems already in use or under development, a key question concerns trust. My central argument is that having trustworthy, well-working systems is not enough. To enable trust, the design of autonomous systems also needs to consider other requirements, including a capacity to explain decisions and to have recourse options when things go wrong. The past few years have seen dramatic advances in the deployment of autonomous systems. These are essentially software systems that make decisions and act on them, with real-world consequences.

Kaggle is an AirBnB for Data Scientists – this is where they spend their nights and weekends. It's a crowd-sourced platform to attract, nurture, train and challenge data scientists from all around the world to solve data science and predictive analytics problems through machine learning. It has over 536,000 active members from 194 countries and it receives close to 150,000 submissions per month. Started from Melbourne, Australia Kaggle moved to Silicon Valley in 2011, raised some 11 million dollars from the likes of Hal Varian (Chief Economist at Google), Max Levchin (Paypal), Index and Khosla Ventures and then ultimately been acquired by the Google in March of 2017. Kaggle is the number one stop for data science enthusiasts all around the world who compete for prizes and boost their Kaggle rankings.

Deep learning neural networks -- the likes of which power AlphaGo as well as the current generation of image recognition and language translation systems -- are the best machine learning systems we've developed to date. They're capable of incredible feats but still face significant technological hurdles, like the fact that in order to be trained on a specific skill they require upfront access to massive data sets. What's more if you want to retrain that neural network to perform a new skill, you've essentially got to wipe its memory and start over from scratch -- a process known as "catastrophic forgetting". Compare that to the human brain, which learns incrementally rather than bursting forth fully-formed from a sea of data points. It's a fundamental difference: deep learning AIs are generated from the top down, knowing everything it needs to from the get-go, while the human mind is built from the ground up with previous lessons learned being applied to subsequent experiences to create new knowledge.

Driverless vehicles could build a "gold mine" of personal data for private companies and would make it easier for them to target people as consumers, an Australian law professor has warned. Des Butler, of the Queensland University of Technology, said the privacy risks involved in driverless vehicles were a "sleeper issue" that regulators were yet to fully consider, even though car manufacturers say the technology could be on roads in Australia by 2020. "These vehicles will know where you like to frequent, which businesses, and may very well build a profile of you," Butler said. "People will go into these things not realising just how much data the vehicle will be generating about them and not knowing the extent to which the data can be used." On Thursday, the federal government formally launched a $55m bid to answer some of the questions that surround the nascent technology.

Generative Adversarial Networks (GANs) produce systematically better quality samples when class label information is provided., i.e. in the conditional GAN setup. This is still observed for the recently proposed Wasserstein GAN formulation which stabilized adversarial training and allows considering high capacity network architectures such as ResNet. In this work we show how to boost conditional GAN by augmenting available class labels. The new classes come from clustering in the representation space learned by the same GAN model. The proposed strategy is also feasible when no class information is available, i.e. in the unsupervised setup. Our generated samples reach state-of-the-art Inception scores for CIFAR-10 and STL-10 datasets in both supervised and unsupervised setup.