Addenbrooke's Hospital in Cambridge and 20 other hospitals from across the world and healthcare technology leader NVIDIA have used artificial intelligence (AI) to predict COVID patients' oxygen needs on a global scale. The research was sparked by the pandemic and set out to build an AI tool to predict how much extra oxygen a COVID-19 patient might need in the first days of hospital care, using data from across four continents. The technique, known as federated learning, used an algorithm to analyze chest X-rays and electronic health data from hospital patients with COVID symptoms. To maintain strict patient confidentiality, the patient data was fully anonymized and an algorithm was sent to each hospital so no data was shared or left its location. Once the algorithm had "learned" from the data, the analysis was brought together to build an AI tool which could predict the oxygen needs of hospital COVID patients anywhere in the world.

Over the last few years, the concept of Artificial Intelligence has become central in different tasks concerning both our daily life and several working scenarios. Among these tasks automated planning has always been central in the AI research community. In particular, this manuscript is focused on a specialized kind of planning known as Multi-agent Epistemic Planning (MEP). Epistemic Planning (EP) refers to an automated planning setting where the agent reasons in the space of knowledge/beliefs states and tries to find a plan to reach a desirable state from a starting one. Its general form, the MEP problem, involves multiple agents who need to reason about both the state of the world and the information flows between agents. To tackle the MEP problem several tools have been developed and, while the diversity of approaches has led to a deeper understanding of the problem space, each proposed tool lacks some abilities and does not allow for a comprehensive investigation of the information flows. That is why, the objective of our work is to formalize an environment where a complete characterization of the agents' knowledge/beliefs interaction and update is possible. In particular, we aim to achieve such goal by defining a new action-based language for multi-agent epistemic planning and to implement an epistemic planner based on it. This solver should provide a tool flexible enough to reason on different domains, e.g., economy, security, justice and politics, where considering others' knowledge/beliefs could lead to winning strategies.

Symbolic learning represents the most straightforward approach to interpretable modeling, but its applications have been hampered by a single structural design choice: the adoption of propositional logic as the underlying language. Recently, more-than-propositional symbolic learning methods have started to appear, in particular for time-dependent data. These methods exploit the expressive power of modal temporal logics in powerful learning algorithms, such as temporal decision trees, whose classification capabilities are comparable with the best non-symbolic ones, while producing models with explicit knowledge representation. With the intent of following the same approach in the case of spatial data, in this paper we: i) present a theory of spatial decision tree learning; ii) describe a prototypical implementation of a spatial decision tree learning algorithm based, and strictly extending, the classical C4.5 algorithm; and iii) perform a series of experiments in which we compare the predicting power of spatial decision trees with that of classical propositional decision trees in several versions, for a multi-class image classification problem, on publicly available datasets. Our results are encouraging, showing clear improvements in the performances from the propositional to the spatial models, which in turn show higher levels of interpretability.

Answer set programming (ASP) is an efficient problem-solving approach, which has been strongly supported both scientifically and technologically by several solvers, ongoing active research, and implementations in many different fields. However, although researchers acknowledged long ago the necessity of epistemic operators in the language of ASP for better introspective reasoning, this research venue did not attract much attention until recently. Moreover, the existing epistemic extensions of ASP in the literature are not widely approved either, due to the fact that some propose unintended results even for some simple acyclic epistemic programs, new unexpected results may possibly be found, and more importantly, researchers have different reasonings for some critical programs. To that end, Cabalar et al. have recently identified some structural properties of epistemic programs to formally support a possible semantics proposal of such programs and standardise their results. Nonetheless, the soundness of these properties is still under debate, and they are not widely accepted either by the ASP community. Thus, it seems that there is still time to really understand the paradigm, have a mature formalism, and determine the principles providing formal justification of their understandable models. In this paper, we mainly focus on the existing semantics approaches, the criteria that a satisfactory semantics is supposed to satisfy, and the ways to improve them. We also extend some well-known propositions of here-and-there logic (HT) into epistemic HT so as to reveal the real behaviour of programs. Finally, we propose a slightly novel semantics for epistemic ASP, which can be considered as a reflexive extension of Cabalar et al.'s recent formalism called autoepistemic ASP.

While propositional Answer Set Programming (ASP) is already NP-hard and therefore powerful enough to express many challenging problems, their specification can be tedious and complicated. Further, there are relevant problems that require higher expressivity or reasoning over data that changes with time. This and the practical usage of ASP gave rise to a need for a simpler, more expressive, and more concise specification language [1, 11]. Thus, ASP was extended in multiple directions. We focus on the following ones: 1. Time Domain (TD): In [5] ASP-semantics were combined with a temporal context resulting in the Logic-based framework for Analytic Reasoning over Streams (LARS). Here, interpretations assign possibly different sets of facts to time points. Accordingly, the input language was extended with operators like, corresponding to existential quantification over time points. Another temporal extension of ASP is Temporal Equilibrium Logic (TEL) [9].

American football is an increasingly popular sport, with a growing audience in many countries in the world. The most watched American football league in the world is the United States' National Football League (NFL), where every offensive play can be either a run or a pass, and in this work we focus on passes. Many factors can affect the probability of pass completion, such as receiver separation from the nearest defender, distance from receiver to passer, offense formation, among many others. When predicting the completion probability of a pass, it is essential to know who the target of the pass is. By using distance measures between players and the ball, it is possible to calculate empirical probabilities and predict very accurately who the target will be. The big question is: how likely is it for a pass to be completed in an NFL match while the ball is in the air? We developed a machine learning algorithm to answer this based on several predictors. Using data from the 2018 NFL season, we obtained conditional and marginal predictions for pass completion probability based on a random forest model. This is based on a two-stage procedure: first, we calculate the probability of each offensive player being the pass target, then, conditional on the target, we predict completion probability based on the random forest model. Finally, the general completion probability can be calculated using the law of total probability. We present animations for selected plays and show the pass completion probability evolution.

An artificial intelligence (AI) tool was able to distinguish, with great accuracy, Parkinson's patients from healthy peers by analyzing short videos of facial expressions, particularly smiles, a small study shows. The predictive accuracy of the new tool was comparable to that of video analysis that uses motor tasks to detect Parkinson's, pinpointing facial expressions as a potential digital, diagnostic biomarker of the disease. This type of biomarker could allow remote diagnosis without the need for personal interaction and extensive testing. This would be particularly relevant in situations such as a pandemic, in cases of reduced mobility, or in underdeveloped countries where few neurologists exist but most people have access to a phone with a camera, researchers noted. The study, "Facial expressions can detect Parkinson's disease: preliminary evidence from videos collected online," was published as a brief communication in the journal npj Digital Medicine.

Echo-sounder data registered by buoys attached to drifting FADs provide a very valuable source of information on populations of tuna and their behaviour. This value increases when these data are supplemented with oceanographic data coming from CMEMS. We use these sources to develop Tuna-AI, a Machine Learning model aimed at predicting tuna biomass under a given buoy, which uses a 3-day window of echo-sounder data to capture the daily spatio-temporal patterns characteristic of tuna schools. As the supervised signal for training, we employ more than 5000 set events with their corresponding tuna catch reported by the AGAC tuna purse seine fleet.

As industrial machines are becoming more connected and flexible, the process of building and commissioning the machine is also getting smarter. Machines are built now using artificial intelligence, digital twins, and augmented reality. We caught up with Rahul Garg, VP of industrial machinery and mid-market program at Siemens Digital Industries Software. Garg explained the process of creating smart industrial machines using advanced technology. Design News: Is artificial intelligence becoming a major factor in building industrial machines?

Robotics is a fast-growing industry. A recent report from GlobalData forecasts that it will pass the $500bn mark in 2030, after a decade of double-digit annual growth. That's an impressive figure for an industry that generated global revenue of just $45bn in 2020. Most of the value generated by robotics comes from service robots, a broad category that includes consumer robots, as well as robots used in logistics, healthcare, security, and many other areas of the service sector. However, industrial robots will grow at a faster rate in the 2020s.