OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games. OpenSpiel supports n-player (single- and multi- agent) zero-sum, cooperative and general-sum, one-shot and sequential, strictly turn-taking and simultaneous-move, perfect and imperfect information games, as well as traditional multiagent environments such as (partially- and fully-observable) grid worlds and social dilemmas. OpenSpiel also includes tools to analyze learning dynamics and other common evaluation metrics. This document serves both as an overview of the code base and an introduction to the terminology, core concepts, and algorithms across the fields of reinforcement learning,computational game theory, and search. Read More
Monthly Archives: August 2019
China’s path to AI domination has a problem: brain drain
A new analysis shows that the number of Chinese AI researchers has increased tenfold over the last decade, but the majority of them live outside the country.
Superpower dreams: China has put forth a concerted effort to grow into a leading AI powerhouse over the last few years. Beijing deemed the discipline in need of special attention as early as 2012, and in 2017 it released a detailed national strategy for advancing and harnessing the technology. Read More
Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
Machine learning has evolved into an enabling technology for a wide range of highly successful applications. The potential for this success to continue and accelerate has placed machine learning (ML) at the top of research, economic and political agendas. Such unprecedented interest is fueled by a vision of ML applicability extending to healthcare, transportation, defense and other domains of great societal importance. Achieving this vision requires the use of ML in safety-critical applications that demand levels of assurance beyond those needed for current ML applications. Our paper provides a comprehensive survey of the state-of-the-art in the assurance of ML, i.e. in the generation of evidence that ML is sufficiently safe for its intended use. The survey covers the methods capable of providing such evidence at different stages of the machine learning lifecycle, i.e. of the complex, iterative process that starts with the collection of the data used to train an ML component for a system, and ends with the deployment of that component within the system. The paper begins with a systematic presentation of the ML lifecycle and its stages. We then define assurance desiderata for each stage, review existing methods that contribute to achieving these desiderata, and identify open challenges that require further research. Read More
Is the Threat of ‘Fake Science’ Real?
In the early 1980s, Soviet intelligence began Operation Infektion—a campaign to erode trust in the U.S. government by orchestrating a series of scientific papers and news articles arguing that the U.S. government created the HIV/AIDS virus. As part of the operation, Soviet intelligence services relied on retired biophysicists Lilli and Jakob Segal, who co-authored with university colleague a 47-page pamphlet attributing the origins of the disease to the U.S. government. The Segals’ report recounted numerous factually accurate aspects of the disease but veered away from reality by attributing the origins of HIV/AIDS to U.S. military experiments on prisoners at Fort Detrick, Maryland. Just two years after publication, the report had received coverage from news organizations in more than 80 countries and contributed to the persistent belief that the U.S. government manufactured HIV/AIDS. Read More
WT2 Instant Translation, does it work?
What is Amazon Go, where is it, and how does it work?
Amazon will be opening more Amazon Go stores in the US and UK during 2019 – the latest rumour is that Amazon has now settled on a London site.
Amazon Go gives you the option to buy your goods from Amazon in person rather than through Amazon.com.
However, unlike other physical shops, it doesn’t have any registers or checkouts. You simply walk in, pick out what you want and walk out. Amazon is calling this a “Just Walk Out” shopping experience. Read More
The Anthropologist of Artificial Intelligence
How do new scientific disciplines get started? For Iyad Rahwan, a computational social scientist with self-described “maverick” tendencies, it happened on a sunny afternoon in Cambridge, Massachusetts, in October 2017. Rahwan and Manuel Cebrian, a colleague from the MIT Media Lab, were sitting in Harvard Yard discussing how to best describe their preferred brand of multidisciplinary research. The rapid rise of artificial intelligence technology had generated new questions about the relationship between people and machines, which they had set out to explore. Rahwan, for example, had been exploring the question of ethical behavior for a self-driving car — should it swerve to avoid an oncoming SUV, even if it means hitting a cyclist? — in his Moral Machine experiment.
“I was good friends with Iain Couzin, one of the world’s foremost animal behaviorists,” Rahwan said, “and I thought, ‘Why isn’t he studying online bots? Why is it only computer scientists who are studying AI algorithms?’
“All of a sudden,” he continued, “it clicked: We’re studying behavior in a new ecosystem.”
Two years later, Rahwan, who now directs the Center for Humans and Machines at the Max Planck Institute for Human Development, has gathered 22 colleagues — from disciplines as diverse as robotics, computer science, sociology, cognitive psychology, evolutionary biology, artificial intelligence, anthropology and economics — to publish a paper in Nature calling for the inauguration of a new field of science called “machine behavior.” Read More
China's hackers are ransacking databases for your health data
In May 2017, the WannaCry ransomware spread around the globe. As the worm locked Windows PCs, the UK’s National Health Service quickly ground to a halt. 19,000 appointments were cancelled, doctor’s couldn’t access patient files and email accounts were taken offline.
But North Korean hackers behind WannaCry didn’t touch one thing: patient data. No personal information was stolen, the NHS has concluded. The cyberattack was purely to cause disruption and an attempt to earn the hermit state some much-needed cash.
The same can’t be said for China. New analysis has indicated that state-sponsored hackers from the country are targetting medical data from the healthcare industry. Research from security firm FireEye, has identified multiple groups with links to China attacking medical systems and databases around the world. These attacks include incidents in 2019, but also date back as far as 2013. Read More
How YouTube Radicalized Brazil
When Matheus Dominguez was 16, YouTube recommended a video that changed his life.
He was in a band in Niterói, a beach-ringed city in Brazil, and practiced guitar by watching tutorials online.
YouTube had recently installed a powerful new artificial intelligence system that learned from user behavior and paired videos with recommendations for others. One day, it directed him to an amateur guitar teacher named Nando Moura, who had gained a wide following by posting videos about heavy metal, video games and, most of all, politics.
In colorful and paranoid far-right rants, Mr. Moura accused feminists, teachers and mainstream politicians of waging vast conspiracies. Mr. Dominguez was hooked.
As his time on the site grew, YouTube recommended videos from other far-right figures. One was a lawmaker named Jair Bolsonaro, then a marginal figure in national politics — but a star in YouTube’s far-right community in Brazil, where the platform has become more widely watched than all but one TV channel.
Last year, he became President Bolsonaro. Read More
When speed kills: Lethal autonomous weapon systems, deterrence and stability
While the applications of artificial intelligence (AI) for militaries are broad, lethal autonomous weapon systems (LAWS) represent one possible usage of narrow AI by militaries. Research and development on LAWS by major powers, middle powers and non-state actors makes exploring the consequences for the security environment a crucial task. This article draws on classic research in security studies and examples from military history to assess the potential development and deployment of LAWS, as well as how they could influence arms races, the stability of deterrence, including strategic stability, the risk of crisis instability and wartime escalation. It focuses on these questions through the lens of two characteristics of LAWS: the potential for increased operational speed and the potential for decreased human control over battlefield choices. It also examines how these issues interact with the large uncertainty parameter associated with potential AI-based military capabilities at present, both in terms of the range of the possible and the opacity of their programming. Read More
Rick's Cafe AI 