An Australian team is using machine learning to tackle the threat of space junk wrecking new satellites. Research to tackle the growing need to find, capture and remove junk from space is advancing at the Australian Institute for Machine Learning in Adelaide, South Australia. Machine Learning for Space director Tat-Jun Chin and his Adelaide-based team have won a $600,000 grant from Australia's SmartSat CRC to continue their work in detecting, tracking and cataloging space junk. SmartSat CRC was established last year to work with the Australian Space Agency based in Adelaide, contributing to the Australian government's goal of tripling the size of the space sector to $12 billion and creating as many as 20,000 jobs by 2030. The space junk project is based on developing a space-based surveillance network and tackling the growing challenge of crowding in space.

Time series models predominantly, over the years, have focussed on individual time series via local models. This changed with the popularisation of deep learning techniques. This was also supported by the increase of temporal data availability, which led to many deep learning-based time series algorithms. Due to their natural temporal ordering, time-series data are present in almost every task that is registered, taking into account some notion of ordering. From electronic health records and human activity recognition to acoustic scene classification and cyber-security, time series is encountered in many real-world applications.

Here's how you can still get a free Windows 10 upgrade You can still use Microsoft's free upgrade tools to install Windows 10 on an old PC running Windows 7 or Windows 8.1. No product key is required, and the digital license says you're activated and ready to go. Like previous years, 2018 featured a bevy of buzzword-laden technologies, but we at ZDNet are fatigued by the never-ending stream of acronyms. With that fatigue in mind, we put together a simple test for the year in technology. What technologies talked about today will actually matter in a decade? Here's a look at the technologies compiled by Larry Dignan, Chris Duckett, Jason Hiner and Steve Ranger from 2018 that will matter as well as a few that simply didn't make the cut.

More and more, across the globe, the effects of global warming are being felt. Global movements like Extinction Rebellion have repeatedly caused disruption by protesting in cities around the world and are symbolic of the growing attention being paid to this important issue. Despite the boom in public awareness, methods to combat the issue have been slow to develop – separating rubbish into recycling and general waste is as far as the majority of households go. More advanced technology, such as solar panels and wind turbines remain out of reach to many due to their high cost and space required for installation. However, another technology may have a far bigger impact in the fight against climate change.

Shamus Rae, CEO of tech start-up Engine B, and Kirstin Gillon, Technical Manager in ICAEW's IT Faculty, consider the progress made by AI within finance. Shamus Rae (SR): I look at AI hitting the profession in three stages. Stage one is the different focus on play. The second stage is about process efficiency, and the third one is driving business model change and the service delivery of a specific service like audit. Different firms are at different stages of using AI.

In this paper we present a novel approach to optimise tactical and strategic decision making in football (soccer). We model the game of football as a multi-stage game which is made up from a Bayesian game to model the pre-match decisions and a stochastic game to model the in-match state transitions and decisions. Using this formulation, we propose a method to predict the probability of game outcomes and the payoffs of team actions. Building upon this, we develop algorithms to optimise team formation and in-game tactics with different objectives. Empirical evaluation of our approach on real-world datasets from 760 matches shows that by using optimised tactics from our Bayesian and stochastic games, we can increase a team chances of winning by up to 16.1\% and 3.4\% respectively.

Decentralized stochastic optimization methods have gained a lot of attention recently, mainly because of their cheap per iteration cost, data locality, and their communication-efficiency. In this paper we introduce a unified convergence analysis that covers a large variety of decentralized SGD methods which so far have required different intuitions, have different applications, and which have been developed separately in various communities. Our algorithmic framework covers local SGD updates and synchronous and pairwise gossip updates on adaptive network topology. We derive universal convergence rates for smooth (convex and non-convex) problems and the rates interpolate between the heterogeneous (non-identically distributed data) and iid-data settings, recovering linear convergence rates in many special cases, for instance for over-parametrized models. Our proofs rely on weak assumptions (typically improving over prior work in several aspects) and recover (and improve) the best known complexity results for a host of important scenarios, such as for instance coorperative SGD and federated averaging (local SGD).

Dynamical systems comprised of autonomous agents arise in many relevant problems such as multi-agent robotics, smart grids, or smart cities. Controlling these systems is of paramount importance to guarantee a successful deployment. Optimal centralized controllers are readily available but face limitations in terms of scalability and practical implementation. Optimal decentralized controllers, on the other hand, are difficult to find. In this paper, we use graph neural networks (GNNs) to learn decentralized controllers from data. GNNs are well-suited for the task since they are naturally distributed architectures. Furthermore, they are equivariant and stable, leading to good scalability and transferability properties. The problem of flocking is explored to illustrate the power of GNNs in learning decentralized controllers.

Knowledge graphs (KGs) store highly heterogeneous information about the world in the structure of a graph, and are useful for tasks such as question answering and reasoning. However, they often contain errors and are missing information. Vibrant research in KG refinement has worked to resolve these issues, tailoring techniques to either detect specific types of errors or complete a KG. In this work, we introduce a unified solution to KG characterization by formulating the problem as unsupervised KG summarization with a set of inductive, soft rules, which describe what is normal in a KG, and thus can be used to identify what is abnormal, whether it be strange or missing. Unlike first-order logic rules, our rules are labeled, rooted graphs, i.e., patterns that describe the expected neighborhood around a (seen or unseen) node, based on its type, and information in the KG. Stepping away from the traditional support/confidence-based rule mining techniques, we propose KGist, Knowledge Graph Inductive SummarizaTion, which learns a summary of inductive rules that best compress the KG according to the Minimum Description Length principle---a formulation that we are the first to use in the context of KG rule mining. We apply our rules to three large KGs (NELL, DBpedia, and Yago), and tasks such as compression, various types of error detection, and identification of incomplete information. We show that KGist outperforms task-specific, supervised and unsupervised baselines in error detection and incompleteness identification, (identifying the location of up to 93% of missing entities---over 10% more than baselines), while also being efficient for large knowledge graphs.

Deep Neural Networks (DNN) represent the state of the art in many tasks. However, due to their overparameterization, their generalization capabilities are in doubt and are still under study. Consequently, DNN can overfit and assign overconfident predictions, as they tend to learn highly oscillating decision thresholds. This has been shown to affect the calibration of the confidences assigned to unseen data. Data Augmentation (DA) strategies have been proposed to overcome some of these limitations. One of the most popular is Mixup, which has shown a great ability to improve the accuracy of these models. Recent work has provided evidence that Mixup also improves the uncertainty quantification and calibration of DNN. In this work, we argue and provide empirical evidence that, due to its fundamentals, Mixup does not necessarily improve calibration. Based on our observations we propose a new loss function that improves the calibration, and also sometimes the accuracy. Our loss is inspired by Bayes decision theory and introduces a new training framework for designing losses for probabilistic modelling. We provide state-of-the-art accuracy with consistent improvements in calibration performance.