In this paper, we propose an interpretable LSTM recurrent neural network, i.e., multi-variable LSTM for time series with exogenous variables. Currently, widely used attention mechanism in recurrent neural networks mostly focuses on the temporal aspect of data and falls short of characterizing variable importance. To this end, our multi-variable LSTM equipped with tensorized hidden states is developed to learn variable specific representations, which give rise to both temporal and variable level attention. Preliminary experiments demonstrate comparable prediction performance of multi-variable LSTM w.r.t. encoder-decoder based baselines. More interestingly, variable importance in real datasets characterized by the variable attention is highly in line with that determined by statistical Granger causality test, which exhibits the prospect of multi-variable LSTM as a simple and uniform end-to-end framework for both forecasting and knowledge discovery.

We developed OmicsMapNet approach to take advantage of existing deep leaning frameworks to analyze high-dimensional omics data as 2-dimensional images. The omics data of individual samples were first rearranged into 2D images in which molecular features related in functions, ontologies, or other relationships were organized in spatially adjacent and patterned locations. Deep learning neural networks were trained to classify the images. Molecular features informative of classes of different phenotypes were subsequently identified. As an example, we used the KEGG BRITE database to rearrange RNA-Seq expression data of TCGA diffuse glioma samples as treemaps to capture the functional hierarchical structure of genes in 2D images. Deep Convolutional Neural Networks (CNN) were derived using tools from TensorFlow to learn the grade of TCGA LGG and GBM samples with relatively high accuracy. The most contributory features in the trained CNN were confirmed in pathway analysis for their plausible functional involvement.

The fundamental problem in short-text classification is \emph{feature sparseness} -- the lack of feature overlap between a trained model and a test instance to be classified. We propose \emph{ClassiNet} -- a network of classifiers trained for predicting missing features in a given instance, to overcome the feature sparseness problem. Using a set of unlabeled training instances, we first learn binary classifiers as feature predictors for predicting whether a particular feature occurs in a given instance. Next, each feature predictor is represented as a vertex $v_i$ in the ClassiNet where a one-to-one correspondence exists between feature predictors and vertices. The weight of the directed edge $e_{ij}$ connecting a vertex $v_i$ to a vertex $v_j$ represents the conditional probability that given $v_i$ exists in an instance, $v_j$ also exists in the same instance. We show that ClassiNets generalize word co-occurrence graphs by considering implicit co-occurrences between features. We extract numerous features from the trained ClassiNet to overcome feature sparseness. In particular, for a given instance $\vec{x}$, we find similar features from ClassiNet that did not appear in $\vec{x}$, and append those features in the representation of $\vec{x}$. Moreover, we propose a method based on graph propagation to find features that are indirectly related to a given short-text. We evaluate ClassiNets on several benchmark datasets for short-text classification. Our experimental results show that by using ClassiNet, we can statistically significantly improve the accuracy in short-text classification tasks, without having to use any external resources such as thesauri for finding related features.

The most recent financial upheavals have cast doubt on the adequacy of some of the conventional quantitative risk management strategies, such as VaR (Value at Risk), in many common situations. Consequently, there has been an increasing need for verisimilar financial stress testings, namely simulating and analyzing financial portfolios in extreme, albeit rare scenarios. Unlike conventional risk management which exploits statistical correlations among financial instruments, here we focus our analysis on the notion of probabilistic causation, which is embodied by Suppes-Bayes Causal Networks (SBCNs); SBCNs are probabilistic graphical models that have many attractive features in terms of more accurate causal analysis for generating financial stress scenarios. In this paper, we present a novel approach for conducting stress testing of financial portfolios based on SBCNs in combination with classical machine learning classification tools. The resulting method is shown to be capable of correctly discovering the causal relationships among financial factors that affect the portfolios and thus, simulating stress testing scenarios with a higher accuracy and lower computational complexity than conventional Monte Carlo Simulations.

One of the critical issues when adopting Bayesian networks (BNs) to model dependencies among random variables is to "learn" their structure. This is a well-known NP-hard problem in its most general and classical formulation, which is furthermore complicated by known pitfalls such as the issue of I-equivalence among different structures. In this work we restrict the investigation to a specific class of networks, i.e., those representing the dynamics of phenomena characterized by the monotonic accumulation of events. Such phenomena allow to set specific structural constraints based on Suppes' theory of probabilistic causation and, accordingly, to define constrained BNs, named Suppes-Bayes Causal Networks (SBCNs). Within this framework, we study the structure learning of SBCNs via extensive simulations with various state-of-the-art search strategies, such as canonical local search techniques and Genetic Algorithms. This investigation is intended to be an extension and an in-depth clarification of our previous works on SBCN structure learning. Among the main results, we show that Suppes' constraints do simplify the learning task, by reducing the solution search space and providing a temporal ordering on the variables, which simplifies the complications derived by I-equivalent structures. Finally, we report on tradeoffs among different optimization techniques that can be used to learn SBCNs.

FinTech Futures met with BNP Paribas Global Markets' Cyril Cottu to discuss AI, bots, the demise of email, its investment in Symphony and a host of other tech topics. All fit within three "mega-trends" that are driving the strategy, investments and skillsets of the bank. Is the financial services sector at a tipping point when it comes to technology? He foresees exponential growth in data, computer power and connectivity, with these "mega-trends" driving rapid change. They are also driving much of the bank's strategy and investment, as reflected in its strides in areas such as artificial intelligence (AI) and its investment last year in cloud-based messaging and collaboration platform, Symphony.

At the end of July, the Allegis Group, a talent solutions provider based in released findings from a survey of more than 300 HR professionals, senior-manager level and above, who reported mixed feelings about AI and its impact on the future of work. The research found that 21 percent view AI as something to be excited about, 17 percent consider it both disrupting and enabling, and a lower number, nine percent, believe AI will displace most jobs in 10 years. This mixed view of AI is not surprising because the technology does more than automate tasks. The person whose role no longer includes a certain repetitive task automated by AI may not necessarily lose their job. "Rather, they may now have new responsibilities that more broadly focus on human capabilities that AI cannot deliver," Rachel Russell, executive director of corporate strategy at Allegis, said.

The Member States agreed to work together on the most important issues raised by Artificial Intelligence, from ensuring Europe's competitiveness in the research and deployment of AI, to dealing with social, economic, ethical and legal questions. The Declaration builds further on the achievements and investments of the European research and business community in AI. AI is already used by citizens daily and facilitates both their personal and professional lives. It can also solve key societal challenges, from sustainable healthcare to climate change and from cybersecurity to sustainable migration. Clearly, the technology is becoming a key driver for economic growth through the digitisation of industry and for society as a whole.

Lately, the DJI Drone Photography Award, make creative use of a drone to explore new photographic possibilities. The drone work is awe-inspiring, letting us consider the world from an alternative perspective. The Sand Castles (Part II) by Markel Redondo focuses on highlighting the Spain's problem from a new perspective. "We live in a society with huge housing issues, where many cannot afford a place to live, yet Spain has more than three million empty homes," said Redondo. Hegen loves exploring the relationship between man and nature and uses aerial photography via drone to document landscapes that have been transformed by human intervention.

The US Army recently announced that it is developing the first drones that can spot and target vehicles and people using artificial intelligence (AI). This is a big step forward. Whereas current military drones are still controlled by people, this new technology will decide who to kill with almost no human involvement. Once complete, these drones will represent the ultimate militarisation of AI and trigger vast legal and ethical implications for wider society. There is a chance that warfare will move from fighting to extermination, losing any semblance of humanity in the process.