Online audio advertising is a particular form of advertising used abundantly in online music streaming services. In these platforms, which tend to host tens of thousands of unique audio advertisements (ads), providing high quality ads ensures a better user experience and results in longer user engagement. Therefore, the automatic assessment of these ads is an important step toward audio ads ranking and better audio ads creation. In this paper we propose one way to measure the quality of the audio ads using a proxy metric called Long Click Rate (LCR), which is defined by the amount of time a user engages with the follow-up display ad (that is shown while the audio ad is playing) divided by the impressions. We later focus on predicting the audio ad quality using only acoustic features such as harmony, rhythm, and timbre of the audio, extracted from the raw waveform. We discuss how the characteristics of the sound can be connected to concepts such as the clarity of the audio ad message, its trustworthiness, etc. Finally, we propose a new deep learning model for audio ad quality prediction, which outperforms the other discussed models trained on hand-crafted features. To the best of our knowledge, this is the first large-scale audio ad quality prediction study.

Future Tense is a partnership of Slate, New America, and Arizona State University that examines emerging technologies, public policy, and society. Photoshopping celebrities' faces onto nude photos has been around for years. Now, new technology has spawned a more sinister upgrade to that practice--and it has online communities scrambling to keep nonconsensual adult content off their platforms. This week, Reddit, Twitter, and Pornhub became the latest social platforms to ban deepfakes--fake porn videos in which artificial intelligence superimposes celebrities' (and other people's) faces onto the bodies of adult film actors. With free and easy-to-use facial recognition technology, users can create videos that are nearly indecipherable from reality.

When Snap first started selling its video-recording glasses, the hype was real. The company had only focused on software before that, and was dipping its toe into hardware with a relatively simple product. Add to that the device's limited availability via sporadic pop-up stores, and Spectacles fever spread rapidly. But hype alone cannot sustain a business. A year and a half later, with 150,000 Spectacles sold and hundreds of thousands reportedly languishing in warehouses, the furor has officially died out, apparently along with Snap's hardware ambitions.

I've been refreshing my CS basics with CTCI and LeetCode, but I don't know how much time I should spend on that versus reading papers and refreshing machine learning knowledge. Should I be able to do backpropagation with ease on a whiteboard? Ideally, I'd like to get involved in research and go back to school eventually.

Some firms in China now use artificial intelligence–powered facial recognition programs to confirm identities. BEIJING--In a gleaming high-rise here in northern Beijing's Haidian district, two hardware jocks in their 20s are testing new computer chips that might someday make smartphones, robots, and autonomous vehicles truly intelligent. The onlooker, Chen Yunji, a 34-year-old computer scientist and founding technical adviser of Cambricon Technologies here, explains that traditional processors, designed decades before the recent tsunami of artificial intelligence (AI) research, "are slow and energy inefficient" at processing the reams of data required for AI. "Even if you have a very good algorithm or application," he says, its usefulness in everyday life is limited if you can't run it on your phone, car, or appliance. "Our goal is to change all lives."

Our identities, whether we are aware of them or not, seem to be in a period of transition. Thus, we could say that we find ourselves between this intermediate step that the sociologist Zygmunt Bauman has already referred to as "liquid currency", and what Eli Pariser described in his book The Filter Bubble. The "filter bubble" concept refers to how the Internet, and for that matter, how the algorithms behind it are in charge of deciding what you read, think, and even buy. The Internet seems to provide us with an identity; however, who provides an identity to the conglomerate of algorithms that make up the network of networks? To understand how algorithms work, it is useful to know what they are, and to view the Internet as something more than a completely harmless or totally evil tool.

Writer-director Sadrac González puts an austere, arty spin on a solid science-fiction premise in "Black Hollow Cage," a time-travel movie more in the mold of Andrei Tarkovsky than "Back to the Future." The film looks stunning and presents some provocative ideas, but González's quietly contemplative approach is numbing.

The future belongs to those who can afford it. This may be virtually true in today's world, where surviving retirement can feel impossible, but it's also the literal premise of Altered Carbon, Netflix's new prestige sci-fi series. Based on Richard K. Morgan's novel of same name, the neo-noir is set several hundred years in the future, when human consciousness has been digitized into microchip-like "stacks" constantly being swapped into and out of various bodies, or "sleeves." This technology, along with innovations like human cloning and artificial intelligence, has given society a quantum leap, but it's also sent socioeconomic stratification into overdrive, creating dire new realities for the poor and incarcerated while simultaneously producing an elite upper-class. Called "Mets"--short for "Methuselahs"--the members of Altered Carbon's 0.001 percent have achieved virtual immortality thanks to vaults of their own cloned sleeves and cloud backups full of their stacks.

There's a race on to get into your ears. And Audible is trying to win it. The Amazon-owned audiobook company is just the latest to launch a new range of audio content, in the form of podcasts. It comes as Google attempts to take its spot by offering audiobooks, and other companies like Spotify launch their own podcasts. Audible is referring to the new podcasts as "Audio Shows". But they work in the same way as podcast – they are cut down into a range of shorter episodes that are intended to be listened to in a series, and fans can subscribe and then receive new episodes when they come out.

The Game Developers Conference (GDC) takes place in San Francisco next month and this year's event includes the GDC's first ever film festival. For three days starting March 19th, the GDC will host a selection of documentary and narrative films focused on the art and culture of video games, and Q&As with the filmmakers will follow most of the screenings. Each day of the festival with be themed. The first day, focused on international works, will feature screenings of Branching Paths, Moleman 4 -- Longplay and the premiere of Heting Chen's Indie Games in China. The second puts the spotlight on webseries with the premiere of the feature-length version of The CheckPoint Series and two work-in-progress screenings of Noclip: Horizon: Zero Dawn and Area 5's Outerlands.