This paper describes a prototype system that integrates social media analysis into the European Flood Awareness System (EFAS). This integration allows the collection of social media data to be automatically triggered by flood risk warnings determined by a hydro-meteorological model. Then, we adopt a multi-lingual approach to find flood-related messages by employing two state-of-the-art methodologies: language-agnostic word embeddings and language-aligned word embeddings. Both approaches can be used to bootstrap a classifier of social media messages for a new language with little or no labeled data. Finally, we describe a method for selecting relevant and representative messages and displaying them back in the interface of EFAS.

Gradient Boosting Decision Tree (GBDT) are popular machine learning algorithms with implementations such as LightGBM and in popular machine learning toolkits like Scikit-Learn. Many implementations can only produce trees in an offline manner and in a greedy manner. We explore ways to convert existing GBDT implementations to known neural network architectures with minimal performance loss in order to allow decision splits to be updated in an online manner and provide extensions to allow splits points to be altered as a neural architecture search problem. We provide learning bounds for our neural network.

Complex-valued representations have demonstrated promising results on modeling relational data, i.e., knowledge graphs. This paper proposes a new knowledge graph embedding method. More concretely, we move beyond standard complex representations, adopting expressive hypercomplex representations for learning representations of entities and relations. Hypercomplex embeddings, or Quaternion embeddings (QuatE), are complex valued embeddings with three imaginary components. Different from standard complex (Hermitian) inner product, latent interdependencies (between all components) are aptly captured via the Hamilton product in Quaternion space, encouraging a more efficient and expressive representation learning process. Moreover, Quaternions are intuitively desirable for smooth and pure rotation in vector space, preventing noise from sheer/scaling operators. Finally, Quaternion inductive biases enjoy and satisfy the key desiderata of relational representation learning (i.e., modeling symmetry, anti-symmetry and inversion). Experimental results demonstrate that QuatE achieves state-of-the-art performance on four well-established knowledge graph completion benchmarks.

The fear of robots coming for your job is one of the many challenges confronting 21st-century workers, but the machines aren't ready to take on every industry just yet. Bridgewater Associates, the massive hedge fund founded by legendary investor Ray Dalio, just released a report on the changing relationship between labour and capital in the US. One of the big factors the Bridgewater authors highlighted was the ongoing rise in automation across industries, which they noted could be a support for corporate profits in the years to come as more efficient robots and software potentially replace slower and error-prone human labour. Bridgewater cited a 2016 report from consulting firm McKinsey & Company that looked at which industries in the US were most susceptible to being automated. The McKinsey report used data from the Department of Labour to estimate how much time workers in various industry sectors spent doing different types of tasks, and which of those tasks could, theoretically, be automated using present technology.

Networks are powerful data structures, but are challenging to work with for conventional machine learning methods. Network Embedding (NE) methods attempt to resolve this by learning vector representations for the nodes, for subsequent use in downstream machine learning tasks. Link Prediction (LP) is one such downstream machine learning task that is an important use case and popular benchmark for NE methods. Unfortunately, while NE methods perform exceedingly well at this task, they are lacking in transparency as compared to simpler LP approaches. We introduce ExplaiNE, an approach to offer counterfactual explanations for NE-based LP methods, by identifying existing links in the network that explain the predicted links. ExplaiNE is applicable to a broad class of NE algorithms. An extensive empirical evaluation for the NE method `Conditional Network Embedding' in particular demonstrates its accuracy and scalability.

The use of deep pretrained bidirectional transformers has led to remarkable progress in learning multi-sentence representations for downstream language understanding tasks (Devlin et al., 2018). For tasks that make pairwise comparisons, e.g. matching a given context with a corresponding response, two approaches have permeated the literature. A Cross-encoder performs full self-attention over the pair; a Bi-encoder performs self-attention for each sequence separately, and the final representation is a function of the pair. While Cross-encoders nearly always outperform Bi-encoders on various tasks, both in our work and others' (Urbanek et al., 2019), they are orders of magnitude slower, which hampers their ability to perform real-time inference. In this work, we develop a new architecture, the Poly-encoder, that is designed to approach the performance of the Cross-encoder while maintaining reasonable computation time. Additionally, we explore two pretraining schemes with different datasets to determine how these affect the performance on our chosen dialogue tasks: ConvAI2 and DSTC7 Track 1. We show that our models achieve state-of-the-art results on both tasks; that the Poly-encoder is a suitable replacement for Bi-encoders and Cross-encoders; and that even better results can be obtained by pretraining on a large dialogue dataset.

The $K$-means algorithm is extended to allow for partitioning of skewed groups. Our algorithm is called TiK-Means and contributes a $K$-means type algorithm that assigns observations to groups while estimating their skewness-transformation parameters. The resulting groups and transformation reveal general-structured clusters that can be explained by inverting the estimated transformation. Further, a modification of the jump statistic chooses the number of groups. Our algorithm is evaluated on simulated and real-life datasets and then applied to a long-standing astronomical dispute regarding the distinct kinds of gamma ray bursts.

A pillar box red electric train connects Paris, Verona and Grenada via Budapest's Liberty Bridge and on to Heidelberg Castle in a 120-hectare fantasy business park dreamt up by the Chinese billionaire Ren Zhengfei. Ren, 74, a former Red Army engineer who founded the telecommunications company Huawei in 1987 and still owns a 1.14% stake, asked the Japanese architect Kengo Kuma to recreate some of Europe's most historic cities. He hoped to inspire an army of 25,000 research and development staff to challenge Apple, Google and Samsung. While its US competitors keep their research facilities on lockdown to prevent corporate espionage (oft allegedly by the Chinese), Huawei is inviting the world's media into its labs and factories in an attempt to dispel the US government's claims that the privately held company is an arm of the Chinese state and that its technology could be used to hack into western governments. US politicians allege that Huawei's forthcoming 5G mobile phone networks could be hacked by Chinese spies to eavesdrop on sensitive phone calls, gain access to counter-terrorist operations – and potentially even kill targets by crashing driverless cars.

Intensive care clinicians are presented with large quantities of patient information and measurements from a multitude of monitoring systems. The limited ability of humans to process such complex information hinders physicians to readily recognize and act on early signs of patient deterioration. We used machine learning to develop an early warning system for circulatory failure based on a high-resolution ICU database with 240 patient years of data. This automatic system predicts 90.0% of circulatory failure events (prevalence 3.1%), with 81.8% identified more than two hours in advance, resulting in an area under the receiver operating characteristic curve of 94.0% and area under the precision-recall curve of 63.0%. The model was externally validated in a large independent patient cohort.

Optimal planning with respect to learned neural network (NN) models in continuous action and state spaces using mixed-integer linear programming (MILP) is a challenging task for branch-and-bound solvers due to the poor linear relaxation of the underlying MILP model. For a given set of features, potential heuristics provide an efficient framework for computing bounds on cost (reward) functions. In this paper, we introduce a finite-time algorithm for computing an optimal potential heuristic for learned NN models. We then strengthen the linear relaxation of the underlying MILP model by introducing constraints to bound the reward function based on the precomputed reward potentials. Experimentally, we show that our algorithm efficiently computes reward potentials for learned NN models, and the overhead of computing reward potentials is justified by the overall strengthening of the underlying MILP model for the task of planning over long-term horizons.