Alphabet subsidiary and precision health company Verily recently announced a breakthrough in its AI drug discovery GPCR research collaboration with Sosei Heptares. A mere six months ago Verily launched the study with Sosei Heptares – a global leader in GPCR structure-based drug design – with an aim to "prioritise protein targets for therapeutic targeting in immune-mediated disease". Now, Verily has announced that early results from its "next generation immune mapping technology" Immune Profiler platform have already identified "more effective therapeutic options against G protein-coupled receptors (GPCR) in autoimmune and other immune-mediated diseases". The companies hope that in the year to come those data targets will be entered for validation, hit generation, and lead selection. With approximately one third of all current FDA-approved drugs targeting GPCRs, Verily/Sosei Heptares are looking to expedite GPCR research within not only immunology, but also gastroenterology and immuno-oncology as well, and the latest data bodes well for future development of therapeutic options in these areas.

Small in size, sensitive of constitution and with only 130 breeding pairs surviving locally in the wild, the lesser spotted eagle of the Oder delta lives up to its name. In Germany, key questions over the country's energy future hang on the question of whether artificial intelligence systems can do a better job of spotting the reclusive animal than birdwatchers do. Lesser spotted eagles (named after the drop-shaped spots on their feathers) are fond of riding thermals over many of the flatlands earmarked for a mass expansion of onshore windfarms by a German government under pressure to compensate for a pending loss of nuclear power, coal plants and Russian gas. Because lesser spotted eagles in mid-flight are unused to vertical obstacles, and keep their eyes focused on mice, lizard or frog-shaped prey below, conservationists say, they are known to occasionally collide with the rotor blades of wind turbines. German researchers list eight dead specimens found in the vicinity of windfarms since 2002, a small but not insignificant number given the species' endangered status in the country.

Digital technologies have led to the application of new-age technologies that operate with minimal human intervention. And while they may heighten productivity and drive growth, any failure can pose a significant challenge for IT and DevOps teams to resolve. An incident or service disruption is an IT manager's worst nightmare. Very often, factors such as cybersecurity breaches, human error, and the accelerated pace of innovation place significant pressure on enterprises' IT infrastructure, leading to system failures and outages impacting the bottom line. According to the ITIC 2021 Hourly Cost of Downtime Survey, 44% of participants (of 1,200 global organizations) said that hourly downtime costs anywhere from $1 million to over $5 million.

Internet of Things (IoT), Artificial Intelligence (AI), Machine learning (ML) and Deep Learning, have been fashionable keywords of 2018 and the hype will continue in the year 2019 as well. In the past few years we have seen exciting technological breakthroughs, these advancements changed the way we live and will impact our lives in future as well. AI is among the hottest topics which are going beyond our imagination. Subram Natarajan, CTO, IBM India/SA mentioned the new developments in cyber security space – "In 2018, there has been an upswing in AI adoption with more startups and enterprises implementing AI applications. Going forward, Trust and Transparency will continue to drive the AI-conversation. As AI systems are increasingly being used to make decisions, it is expected that we should now also be able to explain how decisions are made and whether they are fair and unbiased. We'll begin to reap the benefits of an enhanced focus on trust and transparency, with companies applying new anti-bias techniques, in combination with guidance from in-house and industry ethics advisory groups, in order to make their products and platforms fairer. IBM has just introduced AI OpenScale which makes it possible for businesses for the first time to be able to identify bias' existence in AI applications and automatically mitigate that bias. AI will also see an increased adoption in Cybersecurity, with new tools being developed for predicting and countering cyber attacks precisely."

For many years, machine learning methods have been used in a wide range of fields, including computer vision and natural language processing. While machine learning methods have significantly improved model performance over traditional methods, their black-box structure makes it difficult for researchers to interpret results. For highly regulated financial industries, transparency, explainability, and fairness are equally, if not more, important than accuracy. Without meeting regulated requirements, even highly accurate machine learning methods are unlikely to be accepted. We address this issue by introducing a novel class of transparent and interpretable machine learning algorithms known as generalized gloves of neural additive models. The generalized gloves of neural additive models separate features into three categories: linear features, individual nonlinear features, and interacted nonlinear features. Additionally, interactions in the last category are only local. The linear and nonlinear components are distinguished by a stepwise selection algorithm, and interacted groups are carefully verified by applying additive separation criteria. Empirical results demonstrate that generalized gloves of neural additive models provide optimal accuracy with the simplest architecture, allowing for a highly accurate, transparent, and explainable approach to machine learning.

Recent work in black-box adversarial attacks for NLP systems has attracted much attention. Prior black-box attacks assume that attackers can observe output labels from target models based on selected inputs. In this work, inspired by adversarial transferability, we propose a new type of black-box NLP adversarial attack that an attacker can choose a similar domain and transfer the adversarial examples to the target domain and cause poor performance in target model. Based on domain adaptation theory, we then propose a defensive strategy, called Learn2Weight, which trains to predict the weight adjustments for a target model in order to defend against an attack of similar-domain adversarial examples. Using Amazon multi-domain sentiment classification datasets, we empirically show that Learn2Weight is effective against the attack compared to standard black-box defense methods such as adversarial training and defensive distillation. This work contributes to the growing literature on machine learning safety.

Integer sequences are of central importance to the modeling of concepts admitting complete finitary descriptions. We introduce a novel view on the learning of such concepts and lay down a set of benchmarking tasks aimed at conceptual understanding by machine learning models. These tasks indirectly assess model ability to abstract, and challenge them to reason both interpolatively and extrapolatively from the knowledge gained by observing representative examples. To further aid research in knowledge representation and reasoning, we present FACT, the Finitary Abstraction Comprehension Toolkit.

Innovators are creative people who can conjure the ground-breaking ideas that represent the main engine of innovative organizations. Past research has extensively investigated who innovators are and how they behave in work-related activities. In this paper, we suggest that it is necessary to analyze how innovators behave in other contexts, such as in informal communication spaces, where knowledge is shared without formal structure, rules, and work obligations. Drawing on communication and network theory, we analyze about 38,000 posts available in the intranet forum of a large multinational company. From this, we explain how innovators differ from other employees in terms of social network behavior and language characteristics. Through text mining, we find that innovators write more, use a more complex language, introduce new concepts/ideas, and use positive but factual-based language. Understanding how innovators behave and communicate can support the decision-making processes of managers who want to foster innovation.

The forecasting of credit default risk has been an active research field for several decades. Historically, logistic regression has been used as a major tool due to its compliance with regulatory requirements: transparency, explainability, and fairness. In recent years, researchers have increasingly used complex and advanced machine learning methods to improve prediction accuracy. Even though a machine learning method could potentially improve the model accuracy, it complicates simple logistic regression, deteriorates explainability, and often violates fairness. In the absence of compliance with regulatory requirements, even highly accurate machine learning methods are unlikely to be accepted by companies for credit scoring. In this paper, we introduce a novel class of monotonic neural additive models, which meet regulatory requirements by simplifying neural network architecture and enforcing monotonicity. By utilizing the special architectural features of the neural additive model, the monotonic neural additive model penalizes monotonicity violations effectively. Consequently, the computational cost of training a monotonic neural additive model is similar to that of training a neural additive model, as a free lunch. We demonstrate through empirical results that our new model is as accurate as black-box fully-connected neural networks, providing a highly accurate and regulated machine learning method.

Artificial intelligence (AI) has the potential to greatly improve society, but as with any powerful technology, it comes with heightened risks and responsibilities. Current AI research lacks a systematic discussion of how to manage long-tail risks from AI systems, including speculative long-term risks. Keeping in mind the potential benefits of AI, there is some concern that building ever more intelligent and powerful AI systems could eventually result in systems that are more powerful than us; some say this is like playing with fire and speculate that this could create existential risks (x-risks). To add precision and ground these discussions, we provide a guide for how to analyze AI x-risk, which consists of three parts: First, we review how systems can be made safer today, drawing on time-tested concepts from hazard analysis and systems safety that have been designed to steer large processes in safer directions. Next, we discuss strategies for having long-term impacts on the safety of future systems. Finally, we discuss a crucial concept in making AI systems safer by improving the balance between safety and general capabilities. We hope this document and the presented concepts and tools serve as a useful guide for understanding how to analyze AI x-risk.