Smart speakers (and the virtual assistants they house) offer voice control for so many connected devices it's hard to keep count. Those audio gadgets can also assist with a range of questions -- that is, if they can understand you. The Washington Post took a close look at the performance of Alexa and Google Assistant when it comes to understanding people with strong accents. The results show that while these devices are certainly handy and increasingly popular, there's still a lot of room for improvement. The Facebook CEO covered a range of topics, including its evolving approach to fake news and a comment about Holocaust deniers that warranted a clarification afterwards.

Editors' Note: Industry veteran Arvin Patel is Chief Intellectual Property Officer at TiVo and former chief IPO officer at Technicolor, with previous stints at Rovi and IBM. At TiVo, he oversees the company's annual $400 million IP business and about 6,000 issued patents, as well as its pending applications worldwide. He also leads TiVo's global organization of IP professionals overseeing portfolio development and management, licensing, strategic patent acquisition, policy, and the company's long-term innovation strategy. This year marks the 50th anniversary of 2001: A Space Odyssey, Stanley Kubrick's seminal science fiction film that examined our evolution as human beings and technology's role in our development. Looking back, you can argue that Kubrick was surprisingly prescient in depicting man's relationship with machines as companies like Amazon, Google and Apple continue to invest in AI-related products and services.

Dating back to the 1960s, before the internet, MIT Professor Joseph Weizenbaum created the first ever chatbot. He designed a programme to copy human conversation by pairing scripted responses to text users entered into a computer and thus the chatbot was born. Becoming more conversational since the early days, Chatbots have revolutionised the customer experience with instant, personalised responses. Although they still lack some human aspects and the responses are not always perfect, we are able to see how beneficial they can be. In the past few years we've seen an increase in Artificial Intelligence chatbots across social platforms such as Facebook, when it launched bots for messenger in 2016, to voice experiences with smart speakers, such as the Amazon Alexa.

Last May, my wife and baby and I moved from Austin to Dallas, into an actual house in an actual neighborhood on the east side of the city. We had a few weeks to set up before my wife was going to leave for about three months for job training. Aside from my child and wife, I knew no one in town. We spent a week filling our lives with the things we thought were supposed to go in houses: a sleek water heater that hummed like a cyclotron, new fuses, and a noise machine for our son. We learned about our surroundings, too.

For the third time Netflix organized its Personalization, Recommendation and Search Workshop. It was awesome to get invited for this event during my tech holiday in the San Francisco Bay Area. The experienced data scientists from all over Silicon Valley and beyond made it a knowledge-rich day. Google, Microsoft, Netflix, Spotify, and University of Minnesota shared how to understand and serve your users better. There was one subject that all speakers agreed on: classic matrix factorization (collaborative filtering) reached its expiration date. This includes challenges of multi-armed bandits, an implicit feedback approach, top-N ranking techniques, tyranny of the majority and algorithmic bias. At Netflix almost your whole homepage is personalized: the banner, carousels, order, artwork, text and search. That is why they state that a good recommender system considers: what, how, when and where a title is recommended.

Recently we've seen camera developments from both China and MIT that help us see and take photos around corners, but now you don't need exotic infra red, radar or wifi to spot people through walls, apparently all you need are some easily detectable wireless signals and a dash of AI. Following on from another piece of research that let MIT researchers read peoples emotions using just the WiFi signals from their home routers, another team of researchers at MIT have developed a system, called RF-Pose, where RF stands for Radio Frequency, that uses a neural network to teach RF equipped devices to sense people's movement and postures behind obstacles, and it could be used to help people keep track of elderly relatives in their homes, help gamers turn the house into a giant battleground, and help rescuers rescue people. The team trained their AI to recognise human motion in RF by showing it examples of both on camera movement and signals reflected from people's bodies, helping it understand how the reflections correlate to a given posture. From there the AI could use wireless alone to estimate someone's movements and represent them using stick figures. The scientists mainly see their invention as useful for health care, for the moment anyway, where it could be used to track the development of diseases like Multiple Sclerosis and Parkinson's disease.

Sean Murray does not like talking to the press. He says this several times when we meet at the Guildford offices of Hello Games, the development studio he founded in 2008 with Grant Duncan, Ryan Doyle and David Ream. He is loquacious, but nervous. No one at the studio has spoken to any journalists for nearly two years, since the release of Murray's pet project No Man's Sky, an extraordinarily ambitious space exploration game that aimed to put an infinite universe on a games console – a game that, when it didn't meet some players' high expectations, triggered an appalling internet harassment campaign that left the small studio and its staff reeling. It is hard to blame him for his hesitance.

Human Factors in Hypertext (HUMAN) Opinion Mining, Summarization and Diversification Narrative and Hypertext I attended the Opinion Mining, Summarization and Diversification workshop. The workshop started with a talk titled: "On Reviews, Ratings and Collaborative Filtering," presented by Dr. Oren Sar Shalom, principal data scientist at Intuit, Israel. Next, Ophélie Fraisier, a PhD student studying stance analysis on social media at Paul Sabatier University, France, presented: "Politics on Twitter: A Panorama," in which she surveyed methods of analyzing tweets to study and detect polarization and stances, as well as election prediction and political engagement. He showed how collective opinion mining can help capture the drivers behind opinions as opposed to individual opinion mining (or sentiment) which identifies single individual attitudes toward an item. I thank a million people! https://t.co/I3quPp6nw3 He also discussed a phenomenon in which people are likely to lie to pollsters (social desirability bias) but are honest to Google ("Digital Truth Serum") because Google incentivizes telling the truth. The paper sessions followed the keynote with two full papers and a short paper presentation. Google search data as "digital truth serum" - while reporting of child abuse go down at the recession time, Google search data indicates that real child abuse increases https://t.co/DQQoAotZqB However, it feels more like a research talk rather than a #keynote. Though still interesting, I'd rather hear about a #vision for this area of #research.

A new series of last year's TV hit The Rap of China has kicked off and with it comes the show's first female judge joining the likes of Kris Wu and MC Hotdog. The selection of Hong Kong singer G.E.M., real name Gloria Tang Tsz-Kei, has raised some eyebrows among critics due to the use of artificial intelligence in her selection, as well as her hip-hop credentials. "G.E.M joining The Rap of China is a bit of an embarrassment as her previous image falls short in terms of rap elements," one music critic after learning of the singer's inclusion on the talent show's judging panel. Although the news has drawn widespread criticism from a host of music critics, the move was not unexpected. With G.E.M's involvement, this show hit a new peak in terms of viewing figures in its first week.

ABSTRACT This paper addresses the problem of stylized text generation in a multilingual setup. A version of a language model based on a long short-term memory (LSTM) artificial neural network with extended phonetic and semantic embeddings is used for stylized poetry generation. Phonetics is shown to have comparable importance for the task of stylized poetry generation as the information on the target author. The quality of the resulting poems generated by the network is estimated through bilingual evaluation understudy (BLEU), a survey and a new cross-entropy based metric that is suggested for the problems of such type. The experiments show that the proposed model consistently outperforms random sample and vanilla-LSTM baselines, humans also tend to attribute machine generated texts to the target author. Index Terms-- stylized text generation, poetry generation, artificial neural networks, multilingual models 1. INTRODUCTION The problem of making machine-generated text feel more authentic has a number of industrial and scientific applications, see, for example, [1] or [2]. Most modern generative models are trained on huge corpora of texts which include different contributions from various authors.