A BSTRACT This paper is concerned with the defense of deep models against adversarial attacks. We develop an adversarial detection method, which is inspired by the certificate defense approach, and captures the idea of separating class clusters in the embedding space to increase the margin. The resulting defense is intuitive, effective, scalable, and can be integrated into any given neural classification model. Our method demonstrates state-of-the-art (detection) performance under all threat models. 1 Introduction Defending machine learning models from adversarial attacks has become an increasingly pressing issue as deep neural networks become associated with more critical aspects of society. Adversarial attacks can effectively fool deep models and force them to misclassify, using a slight but maliciously-designed distortion that is typically invisible to the human eye (Carlini & Wagner, 2017c; Athalye et al., 2018). Despite numerous developments, defense mechanisms are still wanting. Many interesting ideas have been proposed to construct defense mechanisms for adversarial examples. Among these are adversarial training (Metzen et al., 2017; Zuo et al., 2020; Y an et al., 2018), ensemble methods (Strauss et al., 2017), and randomization (Dhillon et al., 2018; Xu et al., 2017) to name a few.

Making decisions in the presence of a strategic opponent requires one to take into account the opponent's ability to actively mask its intended objective. To describe such strategic situations, we introduce the non-cooperative inverse reinforcement learning (N-CIRL) formalism. The N-CIRL formalism consists of two agents with completely misaligned objectives, where only one of the agents knows the true objective function. Formally, we model the N-CIRL formalism as a zero-sum Markov game with one-sided incomplete information. Through interacting with the more informed player, the less informed player attempts to both infer, and act according to, the true objective function. As a result of the one-sided incomplete information, the multi-stage game can be decomposed into a sequence of single-stage games expressed by a recursive formula. Solving this recursive formula yields the value of the N-CIRL game and the more informed player's equilibrium strategy. Another recursive formula, constructed by forming an auxiliary game, termed the dual game, yields the less informed player's strategy. Building upon these two recursive formulas, we develop a computationally tractable algorithm to approximately solve for the equilibrium strategies. Finally, we demonstrate the benefits of our N-CIRL formalism over the existing multi-agent IRL formalism via extensive numerical simulation in a novel cyber security setting.

Prec ision Medicine (PM) is an emerging approach that appears with the impression of changing the existing paradigm of medical practice. Recent advances in technological innovations and genetics, and the growing availability of health data have set a new pace o f the research and imposes a set of new requirements on different stakeholders. To date, some studies are available that discuss about different aspects of PM. Nevertheless, a holistic representation of those aspects deemed to confer the technological pers pective, in relation to applications and challenges, is mostly ignored. In this context, this paper surveys advances in PM from informatics viewpoint and reviews the enabling tools and techniques in a categorized manner. In addition, the study discusses ho w other technological paradigms including big data, artificial intelligence, and internet of things can be exploited to advance the potentials of PM. Furthermore, the paper provides some guidelines for future research for seamless implementation and wide - s cale deployment of PM based on identified open issues and associated challenges. To this end, the paper proposes an integrated holistic framework for PM motivating informatics researchers to design their relevant research works in an appropriate context.

Permission to make digital or hard copies of part or all of thi s work for personal or classroom use is granted without fee provided that copies ar e not made or distributed for profit or commercial advantage and that copies bear this n otice and the full citation on the first page. CSCW'19,, November 9th-13th 2019, Austin, T exas ACM 978-1-4503-6819-3/20/04. https://doi.org/10.1145/3334480.XXXXXXX Abstract While machine learning (ML) systems have produced great advances in several domains, their use in support of complex cooperative work remains a research challenge. A particularly challenging setting, and one that may benefit from ML support is the work of multidisciplinary medical teams (MDTs). This paper focuses on the activities performed during the multidisciplinary medical team meeting (MDTM), reviewing their main characteristics in light of a longitud inal analysis of several MDTs in a large teaching hospital over a period of ten years and of our development of ML methods to support MDTMs, and identifying opportunities and possible pitfalls for the use of ML to support MDTMs. Author Keywords Machine Learning; Speech and Language Processing; Mul-tidisciplinary Medical T eam Meeting; Collaboration Introduction An MDT is a group of specialists from different healthcare professions who collaborate on diagnosis and treatment of patients in their care.

LinkedIn cofounder and Greylock Partners investor Reid Hoffman tells executives who are running startups that scale fast -- the kind who want to double in size every few months -- to build ethics into their businesses. As companies plan for the future and grow their engineering or sales ranks, they should consider what can go wrong, he said, and hire people whose job is dedicated to risk management. Next, he added, companies can develop a risk framework to sort risk levels. Anything that can be a catastrophic risk to individuals, a systemic risk to company systems, or a risk to a large number of users should be handled in a proactive way to stay competitive with other startups. Hoffman, who coauthored the book Blitzscaling, joined former White House chief data scientist DJ Patil and Stanford University political science professor Amy Zegart Tuesday at the Stanford Human-Centered AI Intelligence (HAI) fall conference on AI ethics, governance, and policy symposium at the Hoover Institution in Palo Alto.

It's impossible to say precisely how artificial intelligence will disrupt the job market, so researchers at PwC have taken a birds eye view from the top down, and pointed to the results of sweeping economic changes. Their prediction, in a new report out Tuesday, is that it'll all balance out in the end. But the rise in robots and machine-learning software will make the country more productive over the next two decades, growing at a 2% annual clip, to put nearly the same number of jobs back in the system: 7.2 million, PwC estimates. To be clear those new jobs won't involve building robots or coding AI-powered software, which will only make up around 5% of employment, says John Hawksworth, PwC's chief economist. Instead around 1.5 million, or 22%, of the new jobs will be in health and social work.

While many federal agencies primarily think of artificial intelligence as an emerging technology to support their own missions, they also have a secondary role to play in fueling America's research, development and testing of AI by sharing their data, federal tech leaders said Wednesday. The development of innovative artificial intelligence applications relies on powerful underlying data, which many federal agencies hold via the services they provide to Americans. But both U.S. CIO Suzette Kent and Lynne Parker, assistant director of AI in the White House's Office of Science and Technology Policy, identified agencies' hesitancy to share their data with private and academic partners, as well as other agencies, as a leading challenge limiting the nation's development of meaningful AI solutions. Kent said one of her biggest concerns around AI is figuring out "how we make available the powerful data that are strategic assets of the federal government on behalf of its citizens." Agencies are responsible for handling the data properly, but much of it belongs to the public. "The agencies have a responsibility for the external components -- many of the things … around making data available, responding to request from industry, supporting research and development, whether that is in direct grants or specific topic areas or making data or facilities available to support those sets of activities," Kent said at a Bipartisan Policy Center event.

It is incumbent upon business leaders and Australian organisations to put diversity and the ethical implications of artificial intelligence (AI) at the heart of innovation if we're to ensure the world's third major disruptive force is harnessed for human good. That was the big call-out made by Dr Catriona Wallace, founder and executive director of the ASX-listed machine learning tech innovator, Flamingo AI, during this week's CeBIT conference in Sydney. Speaking on the rise of AI and the relationship between humans and machines, the entrepreneur highlighted several facts and figures on the extent of AI impact and innovation over the short and longer-term horizon, as well as the good and negative potential human consequences that come with it. As outlined by Dr Wallace, disruptive technologies, such as AI, are predicted to be the third of three major problems the world is facing that could detrimentally affect humanity. The other two are climate change, and nuclear war.

The Centers for Medicare and Medicare Services this week announced the 25 participants selected to move on to the next round of its Artificial Intelligence Health Outcomes Challenge. WHY IT MATTERS Launched this past March by the CMS Innovation Center, in collaboration with the American Academy of Family Physicians and the Laura and John Arnold Foundation, the AI Health Outcomes Challenge aims to give innovators a showcase for how they're developing AI and machine learning technologies, deep learning tools and neural networks. While the focus is on helping hospitals and health systems drive cost efficiencies for value based reimbursement, prevent adverse patient safety events and boost quality outcomes, CMS put out the call innovators from all sectors of the economy – not just from healthcare. More than 300 different organizations submitted proposals. They were evaluated by a group of data science experts, clinical informaticists and care providers. A CMS selection panel then chose 25 of the applicants to advance to Stage 1.

How do OfSTED determine which schools to inspect? On Wednesday 11th April, I attended an NAHT meeting, a new commission on accountability, spanning every phase and sector of education. Over the next few months it will canvass the views of some of the foremost thinkers in this area of education policy with the aim to have interim findings before the summer term and to publish our full report in September 2018. This post captures a presentation delivered by an OfSTED representative and not the meeting itself. When will [XYZ school] be inspected?