This week's hardware IRL is timely, as Google is about to showcase what's coming next at its annual developer conference. It's the perfect occasion to hear Timothy J. Seppala's thoughts on the Google Home Mini, the company's entry-level smart speaker. He ended up buying two, but they haven't quite delivered on everything he was hoping for. My relationship with a pair of Google Home Mini speakers started out great. I bought them during the holidays as cheap Christmas gifts to myself, hoping to eventually place one in every room of my apartment.

Similar protections don't exist for the two billion-plus Facebook users who don't work for the company, the people said. The dual standard for employees versus regular users is a window on Facebook's struggle over how much to disclose to users about how their data is handled--an issue Facebook has recently tried to address with a raft of changes to the platform. A Facebook spokesman said the company has had discussions about issuing these types of alerts to all users. "In thinking about how we could do something similar for everyone, there are a number of important considerations that come into play--for example, how we can avoid tipping off bad actors or hindering our work to prevent real world harm in cases of abuse or other sensitive situations," the spokesman added. The system can be abused: Earlier this week, Facebook fired a security engineer who had bragged to a woman he met on a dating app about his access to private user information, according to a person familiar with the matter.

Doomsday conspiracies of the human race being ruled by robots aside, Artificial Intelligence (AI) is here to stay. From changing the job landscape, to becoming a buzzword for almost every company, AI is now an integral part of most organisations. Whether it is Google, Facebook, or even companies like Oracle, Microsoft and SAP, they are all working to design software that can learn, and make decisions – in short, Artificial Intelligence. It isn't only private organisations that are keen on the technology, but government organisations like Niti Aayog are also looking closely at it. In fact, the Indian Government has allocated Rs 3,073 crore to spearhead work on fifth generation technology startups like Artificial Intelligence, Machine Learning (ML), Internet of Things (IoT), 3D printing, and Blockchain.

Facebook is adding a dating layer to its main mobile app, CEO Mark Zuckerberg announced today during the company's F8 developer conference keynote in San Jose, California. The features are a long time coming for the 14-year-old social network, which has allowed users to broadcast whether they're single or in a relationship since it first went live in February 2004. The move will likely transform Facebook, with its more than 2.2 billion monthly active users, into a major competitor of Match Group, which owns and operates mobile dating app Tinder and popular dating platform OkCupid. Match Group's stock plummeted by more than 17 percent as soon as the news was announced. Facebook is taking on Tinder with new dating features.

People use artificial intelligence – or AI – any time they ask Siri, Alexa or Google to help them find something. But AI is also changing how health care providers treat patients. "Finding data that's faster, that works continuously like computers do to help make rare diagnoses or faster diagnoses," Dr. David Freeman, a Mayo Clinic neurologist, says. Dr. Freeman has helped develop AI that could soon improve outcomes for people who suffer from a certain kind of stroke called an intracerebral hemorrhage, or ICH. Right now, patients with an ICH go to a hospital with symptoms, get a CAT scan, then have to wait for results and for doctors to figure out how to address it.

Machine learning and artificial intelligence are two of the most exciting and fast-paced trends in tech today, yet despite referring to distinctly separate fields of study, they're often incorrectly used as interchangeable terms. It's best to think of AI being an umbrella term, a fairly wide-ranging categorisation that involves a number of different methodologies and end-goals for the technology. Machine learning sits within this grouping and is itself a fairly outdated term, with research having further splintered into more nuanced areas. As a general concept, the field of AI attempts to create systems that are able to think and behave in a way that a human would. The principles enshrined in the Turing Test, that a system is deemed intelligent is a human is unable to distinguish its behaviour from that of a human, still holds true today.

Conversational AI is reshaping the financial services enterprise. In fact, more than 2 billion people are expected to be using conversational AI by this year, making the need to integrate it into new business applications increasingly pressing. To that end, 40% of virtual assistant users are expected to interact primarily with new applications that support conversational AI.

However, modern customers (Millennials and Generation Z) are not so easy to surprise. So, it is not enough for retailers to keep up with settled trends -- they have to be few steps ahead to hold all the aces. Artificial Intelligence (AI) becomes a life jacket in this case. How can retail benefit from Artificial Intelligence solutions? Much as television in its time or Internet accordingly, Artificial Intelligence is here yet to make a revolution in common things.

User and item features of side information are crucial for accurate recommendation. However, the large number of feature dimensions, e.g., usually larger than 10^7, results in expensive storage and computational cost. This prohibits fast recommendation especially on mobile applications where the computational resource is very limited. In this paper, we develop a generic feature-based recommendation model, called Discrete Factorization Machine (DFM), for fast and accurate recommendation. DFM binarizes the real-valued model parameters (e.g., float32) of every feature embedding into binary codes (e.g., boolean), and thus supports efficient storage and fast user-item score computation. To avoid the severe quantization loss of the binarization, we propose a convergent updating rule that resolves the challenging discrete optimization of DFM. Through extensive experiments on two real-world datasets, we show that 1) DFM consistently outperforms state-of-the-art binarized recommendation models, and 2) DFM shows very competitive performance compared to its real-valued version (FM), demonstrating the minimized quantization loss.

For years, people have used Facebook informally to look for dates. Now Mark Zuckerberg says the platform is starting a dating service. Some experts say the move could invite unwanted solicitation.