Technology is at its most impactful when it is applied to addressing big problems. Perhaps there are no bigger problems than the occurrence of calamities, whether in the form of natural disasters, epidemics, or other catastrophic events. It is in response to seismic events, that fast actions can ameliorate acute conditions and mitigate potentially greater disaster. Such was the case in the wake of the recent super Hurricane Florence which had a massive regional impact when it struck the continental U.S. and devastated the Carolinas with severe flooding and widespread damage. It was in the wake Hurricane Florence that Munich Reinsurance Company, commonly known as Munich Re, stepped into the aftermath to help devastated homeowners and business owners get back on their feet, as it has in response to past calamities.

Can artificial intelligence help us explain why humans do what they do? An international team of cognitive, computer and social scientists has taken tentative steps in this direction by creating a model which uses "psychologically realistic AI" to try to better understand the causes of religious violence and the potential to control it. In simple terms, they combined computer modelling and cognitive psychology to create a multi-agent AI system they say can mimic human religiosity, allowing them to explore the conditions, triggers and patterns for escalating tensions and clashes. The study was built around the question of whether people are naturally violent, or if factors such as religion can cause xenophobic tension and anxiety between different groups that may lead to violence. A team led by Justin Lane from Oxford University in the UK used two real-life examples of "xenophobic social anxiety" that led to extreme physical violence – the 30-year "Troubles" in North Ireland and the three-day Gujurat riots in India in 2002. However, the researchers say the findings can be applied to any occurrence of religious violence, particularly events of radicalised Islam, when people's patriotic identity conflicts with their religious one.

Microsoft has revealed two new training programs to tackle the shortage of AI-related skills in business and academia. What is AI? Everything you need to know about Artificial Intelligence The first of the two programs, Microsoft AI Academy, will run face-to-face and online training sessions for business and public-sector leaders, IT professionals, developers, and startups. "The academy will be helping to develop practical AI skills, learning, and certification for customers and partners," said Cindy Rose, Microsoft UK CEO, speaking at the Future Decoded event in London today. Rose added that Microsoft will use the academy to train up its own staff, including herself. Microsoft's ambition for the academy, she said, is "to empower you and your organization to do more with AI".

Towards a Near Universal Time Series Data Mining Tool: Introducing the Matrix Profile by Chin-Chia Michael Yeh Doctor of Philosophy, Graduate Program in Computer Science University of California, Riverside, September 2018 Dr. Eamonn Keogh, Chairperson The last decade has seen a flurry of research on all-pairs-similarity-search (or, self-join) for text, DNA, and a handful of other datatypes, and these systems have been applied to many diverse data mining problems. Surprisingly, however, little progress has been made on addressing this problem for time series subsequences. In this thesis, we have introduced a near universal time series data mining tool called matrix profile which solves the all-pairssimilarity-search problem and caches the output in an easy-to-access fashion. The proposed algorithm is not only parameter-free, exact and scalable, but also applicable for both single and multidimensional time series. By building time series data mining methods on top of matrix profile, many time series data mining tasks (e.g., motif discovery, discord discovery, shapelet discovery, semantic segmentation, and clustering) can be efficiently solved. Because the same matrix profile can be shared by a diverse set of time series data mining methods, matrix profile is versatile and computed-once-use-many-times data structure. We demonstrate the utility of matrix profile for many time series data mining problems, including motif discovery, discord discovery, weakly labeled time series classification, and vi representation learning on domains as diverse as seismology, entomology, music processing, bioinformatics, human activity monitoring, electrical power-demand monitoring, and medicine. We hope the matrix profile is not the end but the beginning of many more time series data mining projects.

We consider a fair division setting in which $m$ indivisible items are to be allocated among $n$ agents, where the agents have additive utilities and the agents' utilities for individual items are independently sampled from a distribution. Previous work has shown that an envy-free allocation is likely to exist when $m=\Omega(n\log n)$ but not when $m=n+o(n)$, and left open the question of determining where the phase transition from non-existence to existence occurs. We show that, surprisingly, there is in fact no universal point of transition---instead, the transition is governed by the divisibility relation between $m$ and $n$. On the one hand, if $m$ is divisible by $n$, an envy-free allocation exists with high probability as long as $m\geq 2n$. On the other hand, if $m$ is not "almost" divisible by $n$, an envy-free allocation is unlikely to exist even when $m=\Theta(n\log n/\log\log n)$.

Most modern neural machine translation (NMT) systems rely on presegmented inputs. Segmentation granularity importantly determines the input and output sequence lengths, hence the modeling depth, and source and target vocabularies, which in turn determine model size, computational costs of softmax normalization, and handling of out-of-vocabulary words. However, the current practice is to use static, heuristic-based segmentations that are fixed before NMT training. This begs the question whether the chosen segmentation is optimal for the translation task. To overcome suboptimal segmentation choices, we present an algorithm for dynamic segmentation based on the Adaptative Computation Time algorithm (Graves 2016), that is trainable end-to-end and driven by the NMT objective. In an evaluation on four translation tasks we found that, given the freedom to navigate between different segmentation levels, the model prefers to operate on (almost) character level, providing support for purely character-level NMT models from a novel angle.

Gaussian process (GP) audio source separation is a time-domain approach that circumvents the inherent phase approximation issue of spectrogram based methods. Furthermore, through its kernel, GPs elegantly incorporate prior knowledge about the sources into the separation model. Despite these compelling advantages, the computational complexity of GP inference scales cubically with the number of audio samples. As a result, source separation GP models have been restricted to the analysis of short audio frames. We introduce an efficient application of GPs to time-domain audio source separation, without compromising performance. For this purpose, we used GP regression, together with spectral mixture kernels, and variational sparse GPs. We compared our method with LD-PSDTF (positive semi-definite tensor factorization), KL-NMF (Kullback-Leibler non-negative matrix factorization), and IS-NMF (Itakura-Saito NMF). Results show that the proposed method outperforms these techniques.

As statistical classifiers become integrated into real-world applications, it is important to consider not only their accuracy but also their robustness to changes in the data distribution. Although identifying and controlling for confounding variables Z - correlated with both the input X of a classifier and its output Y - has been assiduously studied in empirical social science, it is often neglected in text classification. This can be understood by the fact that, if we assume that the impact of confounding variables does not change between the time we fit a model and the time we use it, then prediction accuracy should only be slightly affected. We show in this paper that this assumption often does not hold and that when the influence of a confounding variable changes from training time to prediction time (i.e. under confounding shift), the classifier accuracy can degrade rapidly. We use Pearl's back-door adjustment as a predictive framework to develop a model robust to confounding shift under the condition that Z is observed at training time. Our approach does not make any causal conclusions but by experimenting on 6 datasets, we show that our approach is able to outperform baselines 1) in controlled cases where confounding shift is manually injected between fitting time and prediction time 2) in natural experiments where confounding shift appears either abruptly or gradually 3) in cases where there is one or multiple confounders. Finally, we discuss multiple issues we encountered during this research such as the effect of noise in the observation of Z and the importance of only controlling for confounding variables.

Graph based entropy, an index of the diversity of events in their distribution to parts of a co-occurrence graph, is proposed for detecting signs of structural changes in the data that are informative in explaining latent dynamics of consumers behavior. For obtaining graph-based entropy, connected subgraphs are first obtained from the graph of co-occurrences of items in the data. Then, the distribution of items occurring in events in the data to these sub-graphs is reflected on the value of graph-based entropy. For the data on the position of sale, a change in this value is regarded as a sign of the appearance, the separation, the disappearance, or the uniting of consumers interests. These phenomena are regarded as the signs of dynamic changes in consumers behavior that may be the effects of external events and information. Experiments show that graph-based entropy outperforms baseline methods that can be used for change detection, in explaining substantial changes and their signs in consumers preference of items in supermarket stores.

Blameworthiness of an agent or a coalition of agents is often defined in terms of the principle of alternative possibilities: for the coalition to be responsible for an outcome, the outcome must take place and the coalition should have had a strategy to prevent it. In this paper we argue that in the settings with imperfect information, not only should the coalition have had a strategy, but it also should have known that it had a strategy, and it should have known what the strategy was. The main technical result of the paper is a sound and complete bimodal logic that describes the interplay between knowledge and blameworthiness in strategic games with imperfect information.