The online analysis of multi-way data stored in a tensor $\mathcal{X} \in \mathbb{R} ^{I_1 \times \dots \times I_N} $ has become an essential tool for capturing the underlying structures and extracting the sensitive features which can be used to learn a predictive model. However, data distributions often evolve with time and a current predictive model may not be sufficiently representative in the future. Therefore, incrementally updating the tensor-based features and model coefficients are required in such situations. A new efficient tensor-based feature extraction, named NeSGD, is proposed for online $CANDECOMP/PARAFAC$ (CP) decomposition. According to the new features obtained from the resultant matrices of NeSGD, a new criteria is triggered for the updated process of the online predictive model. Experimental evaluation in the field of structural health monitoring using laboratory-based and real-life structural datasets show that our methods provide more accurate results compared with existing online tensor analysis and model learning. The results showed that the proposed methods significantly improved the classification error rates, were able to assimilate the changes in the positive data distribution over time, and maintained a high predictive accuracy in all case studies.

Most state-of-the-art speech systems are using Deep Neural Networks (DNNs). Those systems require a large amount of data to be learned. Hence, learning state-of-the-art frameworks on under-resourced speech languages/problems is a difficult task. Problems could be the limited amount of data for impaired speech. Furthermore, acquiring more data and/or expertise is time-consuming and expensive. In this paper we position ourselves for the following speech processing tasks: Automatic Speech Recognition, speaker identification and emotion recognition. To assess the problem of limited data, we firstly investigate state-of-the-art Automatic Speech Recognition systems as it represents the hardest tasks (due to the large variability in each language). Next, we provide an overview of techniques and tasks requiring fewer data. In the last section we investigate few-shot techniques as we interpret under-resourced speech as a few-shot problem. In that sense we propose an overview of few-shot techniques and perspectives of using such techniques for the focused speech problems in this survey. It occurs that the reviewed techniques are not well adapted for large datasets. Nevertheless, some promising results from the literature encourage the usage of such techniques for speech processing.

Current data-driven traffic prediction models are usually trained with large datasets, e.g. several months of speeds and flows. Such models provide very good fit for ordinary road conditions, but often fail just when they are most needed: when traffic suffers a sudden and significant disruption, such as a road incident. In this work, we describe QTIP: a simulation-based framework for quasi-instantaneous adaptation of prediction models upon traffic disruption. In a nutshell, QTIP performs real-time simulations of the affected road for multiple scenarios, analyzes the results, and suggests a change to an ordinary prediction model accordingly. QTIP constructs the simulated scenarios per properties of the incident, as conveyed by immediate distress signals from affected vehicles. Such real-time signals are provided by In-Vehicle Monitor Systems, which are becoming increasingly prevalent world-wide. We experiment QTIP in a case study of a Danish motorway, and the results show that QTIP can improve traffic prediction in the first critical minutes of road incidents.

Exploration is a key problem in reinforcement learning. Recently bonus-based methods have achieved considerable successes in environments where exploration is difficult such as Montezuma's Revenge, which assign additional bonus (e.g., intrinsic reward) to guide the agent to rarely visited states. Since the bonus is calculated according to the novelty of the next state after performing an action, we call such methods the next-state bonus methods. However, the next-state bonus methods bring extra issues. It may lead agent to be trapped in states that fewer being visited and ignore to explore unknown states. Moreover, the behavior policy of the agent is also influenced by the bonus added to the state (or state-action) values indirectly. In contrast to the bonus-based methods which explore in known states, in this paper, we focus on the other part of exploration: exploration for finding unknown states. We propose the action balance exploration method to overcome the defects of the next-state bonus methods, which balances the chosen time of each action in each state and can be treated as an extension of upper confidence bound (UCB) to deep reinforcement learning. To take both the advantages of the next-state bonus method and our action balance exploration method, we propose the action balance RND method, which takes both parts of exploration into consideration. The experiments on grid world and Atari games demonstrate action balance exploration has a better capability in finding unknown states and can improve the real performance of RND in some hard exploration environments respectively.

Currently, we serve pre-trained embeddings for four knowledge graphs. We introduce the service and its usage, and we show further that the trained models have semantic value by evaluating them on multiple semantic benchmarks. The evaluation also reveals that the combination of multiple models can lead to a better outcome than the best individual model.

Computational context understanding refers to an agent's ability to fuse disparate sources of information for decision-making and is, therefore, generally regarded as a prerequisite for sophisticated machine reasoning capabilities, such as in artificial intelligence (AI). Data-driven and knowledge-driven methods are two classical techniques in the pursuit of such machine sense-making capability. However, while data-driven methods seek to model the statistical regularities of events by making observations in the real-world, they remain difficult to interpret and they lack mechanisms for naturally incorporating external knowledge. Conversely, knowledge-driven methods, combine structured knowledge bases, perform symbolic reasoning based on axiomatic principles, and are more interpretable in their inferential processing; however, they often lack the ability to estimate the statistical salience of an inference. To combat these issues, we propose the use of hybrid AI methodology as a general framework for combining the strengths of both approaches. Specifically, we inherit the concept of neuro-symbolism as a way of using knowledge-bases to guide the learning progress of deep neural networks. We further ground our discussion in two applications of neuro-symbolism and, in both cases, show that our systems maintain interpretability while achieving comparable performance, relative to the state-of-the-art.

Perhaps this dog would prefer driving the car, just like three dogs that were trained to do so. We've all seen dogs traveling in cars, including how they like to peek out an open window and enjoy the fur-fluffing breeze and dwell in the cacophony of scents that blow along in the flavorful wind. Dogs have also frequently been used as living props in commercials for cars, pretending in some cases to drive a car, such as the Subaru "Barkleys" advertising campaign that initially launched on TV in 2018 and continued in 2019, proclaiming that Subaru cars were "officially" dog tested and dog approved. What you might not know or might not remember is that there were three dogs that were trained on driving a car and had their moment of unveiling in December of 2012 when they were showcased by driving a car on an outdoor track (the YouTube posted video has amassed millions of views). Yes, three dogs named Monty, Ginny, and Porter were destined to become the first true car drivers on behalf of the entire canine family. Monty at the time was an 18-month-old giant schnauzer cross, while the slightly younger Ginny at one year of age was a beardie whippet cross, and Porter was a youthful 10-month-old beardie.

Artificial intelligence (AI) has the potential to solve many routine business challenges -- from quickly spotting a few questionable charges in thousands of invoices to predicting consumers' needs and wants. But there may be a flipside to these advances. Privacy concerns are cropping up as companies feed more and more consumer and vendor data into advanced, AI-fuelled algorithms to create new bits of sensitive information, unbeknownst to affected consumers and employees. This means that AI may create personal data. When it does, "it's data that has not been provided with [an individual's] consent or even with knowledge", said Chantal Bernier, assistant and interim privacy commissioner in the Office of the Privacy Commissioner of Canada from 2008 until 2014 who now consults in the privacy and cybersecurity practice of global law firm Dentons.

Artificial intelligence (AI) has the potential to solve many routine business challenges -- from quickly spotting a few questionable charges in thousands of invoices to predicting consumers' needs and wants. But there may be a flipside to these advances. Privacy concerns are cropping up as companies feed more and more consumer and vendor data into advanced, AI-fuelled algorithms to create new bits of sensitive information, unbeknownst to affected consumers and employees. This means that AI may create personal data. When it does, "it's data that has not been provided with [an individual's] consent or even with knowledge", said Chantal Bernier, assistant and interim privacy commissioner in the Office of the Privacy Commissioner of Canada from 2008 until 2014 who now consults in the privacy and cybersecurity practice of global law firm Dentons.

There is a gender gap in artificial intelligence (AI). A study by the World Economic Forum and LinkedIn found that only 22% of AI professionals are women. Research by the AI Now Institute found that women make up only 15% of the AI research staff at Facebook and only 10% at Google. Although the gender gap in AI echoes those in cybersecurity and information technology in general, the repercussions of a lack of diversity in AI broaden because the details of the how the systems work are not fully known. As a result, identifying and correcting bias introduced by the decisions of the development teams or the data they select to train their algorithms is difficult.