A self-described "moderate liberal," he also found himself making peace with the conservatism of rural America at a time when those two sides of the US were seemingly not on speaking terms with one another. "It makes you appreciate the commonalities, the things that you can relate to with people from different backgrounds and political views," he says.

A music mashup combines audio elements from two or more songs to create a new work. To reduce the time and effort required to make them, researchers have developed algorithms that predict the compatibility of audio elements. Prior work has focused on mixing unaltered excerpts, but advances in source separation enable the creation of mashups from isolated stems (e.g., vocals, drums, bass, etc.). In this work, we take advantage of separated stems not just for creating mashups, but for training a model that predicts the mutual compatibility of groups of excerpts, using self-supervised and semi-supervised methods. Specifically, we first produce a random mashup creation pipeline that combines stem tracks obtained via source separation, with key and tempo automatically adjusted to match, since these are prerequisites for high-quality mashups. To train a model to predict compatibility, we use stem tracks obtained from the same song as positive examples, and random combinations of stems with key and/or tempo unadjusted as negative examples. To improve the model and use more data, we also train on "average" examples: random combinations with matching key and tempo, where we treat them as unlabeled data as their true compatibility is unknown. To determine whether the combined signal or the set of stem signals is more indicative of the quality of the result, we experiment on two model architectures and train them using semi-supervised learning technique. Finally, we conduct objective and subjective evaluations of the system, comparing them to a standard rule-based system.

In this work, we propose a novel deep learning framework that can generate a vivid dance from a whole piece of music. In contrast to previous works that define the problem as generation of frames of motion state parameters, we formulate the task as a prediction of motion curves between key poses, which is inspired by the animation industry practice. The proposed framework, named DanceNet3D, first generates key poses on beats of the given music and then predicts the in-between motion curves. DanceNet3D adopts the encoder-decoder architecture and the adversarial schemes for training. The decoders in DanceNet3D are constructed on MoTrans, a transformer tailored for motion generation. In MoTrans we introduce the kinematic correlation by the Kinematic Chain Networks, and we also propose the Learned Local Attention module to take the temporal local correlation of human motion into consideration. Furthermore, we propose PhantomDance, the first large-scale dance dataset produced by professional animatiors, with accurate synchronization with music. Extensive experiments demonstrate that the proposed approach can generate fluent, elegant, performative and beat-synchronized 3D dances, which significantly surpasses previous works quantitatively and qualitatively.

I am interested in the best practices for developing computer vision applications for production. Are there tools you would recommend for testing the system? Or do you have to write all the infra in-house?

In our JournalismAI report, journalists around the world told researchers they are eager to collaborate and explore the benefits of AI, especially as it applies to newsgathering, production and distribution. To facilitate their collaboration, the Google News Initiative and Polis – the journalism think tank at the London School of Economics and Political Science – are launching the JournalismAI Collab Challenges, an opportunity for three groups of five newsrooms from the Americas, Europe, the Middle East and Africa, and Asia Pacific to experiment together. Each cohort – selected by Polis – will have six months to either cover global news stories using AI-powered storytelling techniques or to develop prototypes of new AI-based products and processes. Participants will receive support from the JournalismAI team and partner institutions in each region: in the Americas, the challenge will be co-hosted with the Knight Lab at Northwestern University; in Europe, the Middle East and Africa, the challenge will be co-hosted with BBC News Labs and Clwstwr. JournalismAI's partner in Asia Pacific will be announced later this year.

DataIntelo has included a latest report on the Global Artificial Intelligence Products Market into its archive of market research studies.

Artificial intelligence (AI) has been making great strides in recent years … In tests of Project Debater, the AI was given only 15 minutes to research …

As the editor-in-chief of The Verge, I can theoretically assign whatever I want. However, there is one topic I have failed to get people at The Verge to write about for years: robotic process automation, or RPA. RPA isn't robots in factories, which is often what we think of when it comes to automation. This is different: RPA is software. Software that uses other software, like Excel or an Oracle database. On this week's Decoder, I finally found someone who wants to talk about it with me: New York Times tech columnist Kevin Roose. His new book, Futureproof: 9 Rules for Humans in the Age of Automation, has just come out, and it features a lengthy discussion of RPA, who's using it, who it will affect, and how to think about it as you design your career. What struck me during our conversation were the jobs that Kevin talks about as he describes the impact of automation: they're not factory workers and truck drivers. If you have the kind of job that involves sitting in front of a computer using the same software the same way every day, automation is coming for you. It won't be cool or innovative or even work all that well -- it'll just be cheaper, faster, and less likely to complain. That might sound like a downer, but Kevin's book is all about seeing that as an opportunity. You'll see what I mean. Okay, Kevin Roose, tech columnist, author, and the only reporter who has ever agreed to talk to me about RPAs. This transcript has been lightly edited for clarity. Kevin Roose, you're a tech columnist at The New York Times and you have a new book, Futureproof: 9 Rules for Humans in the Age of Automation, which is out now. Thank you for having me. You're ostensibly here to promote your book, which is great. But there's one piece of the book that I am absolutely fascinated by, which is this thing called "robotic process automation." And I'm gonna do my best with you on this show, today, to make that super interesting. But before we get there, let's talk about your book for a minute. What is your book about? Because I read it, and it has a big idea and then there's literally nine rules for regular people to survive. So, tell me how the book came together. So, the book is basically divided into two parts.

There has been lots of buzz around the topic of machine learning (ML) in recent years. Although this concept isn't really new, today, it is literally …

Reinforcement learning is essentially an AI discipline in the sense that it tries to realize content that normally can only be realized by utilizing human …