In this episode of the Data Exchange I speak with Xinyi Zhou, a graduate student in Computer and Information Science at Syracuse University. Xinyi and her advisor (Reza Zafarani) recently wrote a comprehensive survey paper entitled "A Survey of Fake News: Fundamental Theories, Detection Methods, and Opportunities". They set out to organize the many different methods and perspectives used to detect fake news. Their paper is a great resource for anyone wanting to understand the strengths and limitations of various state-of-the-art techniques, and a feel for where the research community might be headed in the near future. We first discussed good working definitions for "fake news".

Two years ago, the Google News Initiative partnered with the London School of Economics and Political Science to launch JournalismAI, a global effort to foster media literacy in newsrooms through research, training and experimentation. Since then, more than 62 thousand journalists have taken Introduction to Machine Learning, an online course provided in 17 languages in partnership with Belgian broadcaster VRT. More than 4,000 people have downloaded the JournalismAI report, which argued that "robots are not going to take over journalism" and that media organizations are keen to collaborate with one another and with technology companies. And over 20 media organizations including La Nación, Reuters, the South China Morning Post and The Washington Post have joined Collab, a global partnership to experiment with AI. To mark this anniversary, together with the London School of Economics, we are hosting a week-long online event to bring together international academics, publishers and practitioners.

Some people's lives are like horror movies. It's strange that, in an age that can create virtual reality, self-driving cars and intelligent machines, the world's third-largest economy can't solve the problem of human misery. More and more Japanese women seem to feel it is. Female suicide is sharply rising. National Police Agency statistics tell the tale, as far as numbers can tell it -- 651 women are known to have taken their own lives that month, up from 400-500 a month typically.

Emily, your lamentation that we don't take the time to discuss big, knotty shows in detail anymore got me thinking. And what I think is … I am to blame? I'm not egomaniacal enough to think I alone am to blame. Streaming platforms that drop entire seasons on as at once might be more responsible, but I too play my tiny part in this state of affairs. The first way is: I like it.

"The tests demonstrated our success in applying artificial intelligence algorithms to'teach' the aircraft's brain to understand what is required of it," says …

The United States' Special Operations Command is currently playing a critical role advancing the employment of AI and machine learning in the fight …

It also requires computer algorithms and artificial intelligence. The United States' Special Operations Command is currently playing a critical role …

As artificial intelligence, robotic process automation and machine learning become more prominent, there is a lot of misinformation and fear about what these technologies actually do and if they will replace people at work. But instead of taking away people's jobs, according to some, AI can help people do their work better. And this is a huge issue -- since more and more people are less engaged at work and turnover rates are peaking in ways never experienced before. Shelly Kramer (pictured), founder and chief executive officer of V3 Broadsuite, president of Broadsuite Media Group and founder and principal analyst at Futurum Research, often writes and speaks about this subject of the relationship between automation and people. "Technology is fueling our world, our personal lives, our business world," Kramer said.

Graph Neural Networks (GNNs) are a framework for graph representation learning, where a model learns to generate low dimensional node embeddings that encapsulate structural and feature-related information. GNNs are usually trained in an end-to-end fashion, leading to highly specialized node embeddings. However, generating node embeddings that can be used to perform multiple tasks (with performance comparable to single-task models) is an open problem. We propose a novel meta-learning strategy capable of producing multi-task node embeddings. Our method avoids the difficulties arising when learning to perform multiple tasks concurrently by, instead, learning to quickly (i.e. with a few steps of gradient descent) adapt to multiple tasks singularly. We show that the embeddings produced by our method can be used to perform multiple tasks with comparable or higher performance than classically trained models. Our method is model-agnostic and task-agnostic, thus applicable to a wide variety of multi-task domains.

Michael Kanaan, AI and machine learning in suicide prevention … He is currently the director of operations for Air Force / MIT artificial intelligence.