Over the past decade, there's been no doubt that AI is positively impacting a number of industries, with medicine and healthcare being no exception. The use of AI and machine learning is already transforming many areas of the healthcare industry, ranging from patient-facing and customer service activities to improvements in overall care, diagnosis, and treatment. With the global pandemic being front-and-center in the minds of healthcare workers, pharmaceutical companies, and life sciences organizations around the world, AI has been applied to help physicians evaluate COVID-19-associated prognosis and needs. In particular, the US federal government has focused on AI in a variety of ways to address pressing needs. The Department of Veterans Affairs, a US Federal agency that provides healthcare services to eligible military veterans, and is the largest integrated health care system in the United States, is adopting AI to help address the wide ranging impacts of the global pandemic on the lives of patients and their families.

The Commonwealth Scientific and Industrial Research Organisation (CSIRO) has announced partnerships with Microsoft, Hobart City Council, and Chemistry Australia to address -- and attempt to end -- Australia's plastics waste issue. Under its plastics mission, CSIRO will work with its partners to develop new solutions that use artificial intelligence (AI), machine learning (ML), and camera sensors for plastics detection and waste monitoring in waterways. CSIRO senior principal research scientist Denise Hardesty said the goal would be to apply technology to the entire plastics supply chain to eliminate rubbish ending up in the environment. "Our research is helping to understand the extent of plastic pollution in Australia and globally, and how to reduce it," she said. "Rethinking plastic packaging is just one way of reducing waste, through better design, materials, and logistics. We can also transform the way we use, manufacture, and recycle plastics by creating new products and more value for plastics."

Machine Learning (ML) algorithms leverage large datasets to determine patterns and construct meaningful recommendations. Likewise, credit risk modelling is a field with access to a large amount of diverse data where ML can be deployed to add analytical value. In the following analysis, we explore how various ML techniques can be used for assessing probability of default (PD) and compare their performance in a real-world setting. A recent publication by the Bank of England (BoE) and the Financial Conduct Authority (FCA) reports the results of a survey on the use of ML in United Kingdom (UK) financial services.[1] Results show that two-thirds of respondents use ML in some form.

Timnit Gebru, a co-leader of Google's Ethical Artificial Intelligence team, said late Wednesday that she'd been abruptly fired from the search giant over an email she sent to colleagues. Gebru, a prominent AI researcher, has done studies on the dangers of facial recognition bias and has spoken out on the lack of diversity in the tech industry. In the wake of her departure, more than 1,400 Google workers, as well as more than 1,800 other industry professionals, signed an open letter in support of Gebru. "Instead of being embraced by Google as an exceptionally talented and prolific contributor, Dr. Gebru has faced defensiveness, racism, gaslighting, research censorship, and now a retaliatory firing," the letter read. In a series of tweets Wednesday night, Gebru said she was terminated for a message she sent to Google Brain Women and Allies, an internal email list at the company.

Gartner's latest Information Security and Risk Management forecast predicts the market will achieve ... [ ] an 8.3% Compound Annual Growth Rate (CAGR) growth rate from 2019 through 2024, reaching $211.4 billion. Bottom Line: In 2021, cybersecurity vendors will accelerate AI and machine learning app development to combine human and machine insights so they can out-innovate attackers intent on escalating an AI-based arms race. Attackers and cybercriminals capitalized on the chaotic year by attempting to breach a record number of enterprise systems in e-commerce, financial services, healthcare and many other industries. AI and machine learning-based cybersecurity apps and platforms combined with human expertise and insights make it more challenging for attackers to succeed in their efforts. Accustomed to endpoint security systems that rely on passwords alone, admin accounts that don't have fundamental security in place, including Multi-Factor Authentication (MFA) and more and attackers created a digital pandemic this year. Interested in what the leading cybersecurity experts are thinking will happen in 2021, I contacted twenty of them who are actively researching how AI can improve cybersecurity next year. Leading experts in the field include including Nicko van Someren, Ph.D. and Chief Technology Officer at Absolute Software, BJ Jenkins, President and CEO of Barracuda Networks, Ali Siddiqui, Chief Product Officer and Ram Chakravarti, Chief Technology Officer, both from BMC, Dr. Torsten George, Cybersecurity Evangelist at Centrify, Tej Redkar, Chief Product Officer at LogicMonitor, Brian Foster, Senior Vice President Product Management at MobileIron, Dr. Mike Lloyd, CTO at RedSeal and many others.

Airbus AI researchers have developed a system that uses natural language understanding to improve question answering (QA) performance when flight crews search for aircraft operating information. The aerospace industry relies on technical documents such as Aircraft Operating Manuals (AOM), Aircraft Operating Instructions and particularly Flight Crew Operating Manuals (FCOM) to guide flight crews on aircraft operations under normal, abnormal, and emergency conditions. FCOMs are issued by aircraft manufacturers and cover system descriptions, procedures, techniques, and performance data. They are the references used to develop standard operating procedures to improve safety and efficiency. Most government aviation administrations have authorized the use of tablet computers by commercial carrier pilots and flight crews to access FCOM information. The Airbus AI researchers note however that existing electronic flight bag (EFB) systems used for this purpose are in practice little more than pdf viewers with keyword search functionality.

Finding the most likely path to a set of failure states is important to the analysis of safety-critical systems that operate over a sequence of time steps, such as aircraft collision avoidance systems and autonomous cars. In many applications such as autonomous driving, failures cannot be completely eliminated due to the complex stochastic environment in which the system operates. As a result, safety validation is not only concerned about whether a failure can occur, but also discovering which failures are most likely to occur. This article presents adaptive stress testing (AST), a framework for finding the most likely path to a failure event in simulation. We consider a general black box setting for partially observable and continuous-valued systems operating in an environment with stochastic disturbances. We formulate the problem as a Markov decision process and use reinforcement learning to optimize it. The approach is simulation-based and does not require internal knowledge of the system, making it suitable for black-box testing of large systems. We present different formulations depending on whether the state is fully observable or partially observable. In the latter case, we present a modified Monte Carlo tree search algorithm that only requires access to the pseudorandom number generator of the simulator to overcome partial observability. We also present an extension of the framework, called differential adaptive stress testing (DAST), that can find failures that occur in one system but not in another. This type of differential analysis is useful in applications such as regression testing, where we are concerned with finding areas of relative weakness compared to a baseline. We demonstrate the effectiveness of the approach on an aircraft collision avoidance application, where a prototype aircraft collision avoidance system is stress tested to find the most likely scenarios of near mid-air collision.

In the last years, AI safety gained international recognition in the light of heterogeneous safety-critical and ethical issues that risk overshadowing the broad beneficial impacts of AI. In this context, the implementation of AI observatory endeavors represents one key research direction. This paper motivates the need for an inherently transdisciplinary AI observatory approach integrating diverse retrospective and counterfactual views. We delineate aims and limitations while providing hands-on-advice utilizing concrete practical examples. Distinguishing between unintentionally and intentionally triggered AI risks with diverse socio-psycho-technological impacts, we exemplify a retrospective descriptive analysis followed by a retrospective counterfactual risk analysis. Building on these AI observatory tools, we present near-term transdisciplinary guidelines for AI safety. As further contribution, we discuss differentiated and tailored long-term directions through the lens of two disparate modern AI safety paradigms. For simplicity, we refer to these two different paradigms with the terms artificial stupidity (AS) and eternal creativity (EC) respectively. While both AS and EC acknowledge the need for a hybrid cognitive-affective approach to AI safety and overlap with regard to many short-term considerations, they differ fundamentally in the nature of multiple envisaged long-term solution patterns. By compiling relevant underlying contradistinctions, we aim to provide future-oriented incentives for constructive dialectics in practical and theoretical AI safety research.

While there has been a surge of recent interest in learning differential equation models from time series, methods in this area typically cannot cope with highly noisy data. We break this problem into two parts: (i) approximating the unknown vector field (or right-hand side) of the differential equation, and (ii) dealing with noise. To deal with (i), we describe a neural network architecture consisting of tensor products of one-dimensional neural shape functions. For (ii), we propose an alternating minimization scheme that switches between vector field training and filtering steps, together with multiple trajectories of training data. We find that the neural shape function architecture retains the approximation properties of dense neural networks, enables effective computation of vector field error, and allows for graphical interpretability, all for data/systems in any finite dimension $d$. We also study the combination of either our neural shape function method or existing differential equation learning methods with alternating minimization and multiple trajectories. We find that retrofitting any learning method in this way boosts the method's robustness to noise. While in their raw form the methods struggle with 1% Gaussian noise, after retrofitting, they learn accurate vector fields from data with 10% Gaussian noise.

Volkswagen recently interviewed a former German constitutional judge, Professor Udo di Fabio. When asked about how Germany's constitution would treat autonomous vehicles, he offered some insight, including both allowing autonomous vehicles and whether manual driving will be allowed after the technology matures. One key quote, which this article will put in context: "People should be able to decide whether to surrender the steering wheel or not." Before I go on, I need to apologize to any German readers, because the rest of us need some key background that you probably got in school. Feel free to skip ahead if you are already familiar with Germany's constitution and its history.