This paper describes a flexible framework for generalized low-rank tensor estimation problems that includes many important instances arising from applications in computational imaging, genomics, and network analysis. The proposed estimator consists of finding a low-rank tensor fit to the data under generalized parametric models. To overcome the difficulty of non-convexity in these problems, we introduce a unified approach of projected gradient descent that adapts to the underlying low-rank structure. Under mild conditions on the loss function, we establish both an upper bound on statistical error and the linear rate of computational convergence through a general deterministic analysis. Then we further consider a suite of generalized tensor estimation problems, including sub-Gaussian tensor denoising, tensor regression, and Poisson and binomial tensor PCA. We prove that the proposed algorithm achieves the minimax optimal rate of convergence in estimation error. Finally, we demonstrate the superiority of the proposed framework via extensive experiments on both simulated and real data.

The time it takes for a classifier to make an accurate prediction can be crucial in many behaviour recognition problems. For example, an autonomous vehicle should detect hazardous pedestrian behaviour early enough for it to take appropriate measures. In this context, we compare the switching linear dynamical system (SLDS) and a three-layered bi-directional long short-term memory (LSTM) neural network, which are applied to infer pedestrian behaviour from motion tracks. We show that, though the neural network model achieves an accuracy of 80%, it requires long sequences to achieve this (100 samples or more). The SLDS, has a lower accuracy of 74%, but it achieves this result with short sequences (10 samples). To our knowledge, such a comparison on sequence length has not been considered in the literature before. The results provide a key intuition of the suitability of the models in time-critical problems.

At the G7 meeting in Montreal last year, Justin Trudeau told WIRED he would look into why more than 100 African artificial intelligence researchers had been barred from visiting that city to attend their field's most important annual event, the Neural Information Processing Systems conference, or NeurIPS. Now the same thing has happened again. More than a dozen AI researchers from African countries have been refused visas to attend this year's NeurIPS, to be held next month in Vancouver. This means an event that shapes the course of a technology with huge economic and social importance will have little input from a major portion of the world. The conference brings together thousands of researchers from top academic institutions and companies, for hundreds of talks, workshops, and side meetings at which new ideas and theories are hashed out. Tejumade Afonja, a master's student from Nigeria who is studying at Saarland University in Germany, posted her rejection letter to Twitter.

With my concept of The Matrix Conspiracy I put myself in the risk of being accused of being a paranoid conspiracy theorist. This is not the case. I m just making aware of that there exists a conspiracy theory which is called The Matrix Conspiracy, and that this conspiracy in fact is a global spreading ideology. My critique is in that way ideology critique, or cultural critique. The concept of the Matrix comes from mathematics, but is more popular known from the movie the Matrix, which asks the question whether we might live in a computer simulation. In The Matrix though, there is also an evil demon, or evil demons, namely the machines which keep the humans in tanks linked to black cable wires that stimulates the virtual reality of the Matrix. Doing this the machines can use the human bodies as batteries that supply the machines with energy. It is the fascination of the virtual reality that deceives the humans. The philosophy behind the movie comes from especially two philosophers: Rene Descartes and George Berkeley. Descartes was very dubious concerning how much we can trust our senses. Therefore he took up the question Is life a dream? However, his intention with this was in his Meditations to develop a confident cognition-argument. In his Meditations Descartes presents the problem approximately like this: I frequently dream during the night, and while I dream, I am convinced, that what I dream is real. But then it always happens, that I wake up and realize, that everything I dreamt was not real, but only an illusion. And then is it I think: is it possible, that what I now, while I am awake, believe is real, also is something, which only is being dreamt by me right now? If it is not the case, how shall I then determinate it? Precisely because Descartes not even in dreams can doubt, that 2 plus 3 is 5, he leaves the dream-argument in his Meditations and goes in tackle with the question, whether he could be cheated by an evil demon concerning all cognition, also the mathematics. This radical skepticism leads him forward to the cogito-argument: Cogito ergo Sum (I think, therefore I exist). But he didn t deny the existence of the external world. The external world he described in a way that resembles what would later be known as modern natural sciences. In the view of nature in natural science, nature is reduced to atomic particles, empty space, fields, electromagnetic waves and particles etc., etc. I have called this the instrumental view of nature. Berkeley is famous for the sentence Esse est percipi, which means that being, or reality, consists in being percepted (to be is to be experienced). The absurdity in Berkeley s assertion is swiftly seen: If a thing, or a human being for that matter, is not being perceived by the senses, then it does not exist. In accordance with Berkeley there therefore does not exist any sense-independent world. He ends in solipsism, the consequence that only I, and my perceptions, can be said to exist.

The global affective computing market is envisioned to create high growth prospects on the back of the rising deployment of machine and human interaction technologies. With enabling technologies already making a mark with their adoption in a range of industry verticals, it could be said that the market has started to evolve. Facial feature extraction software collecting a handsome demand in the recent years is expected to augur well for the growth of the deployment of cameras in affective computing systems. Detection of psychological disorders, facial expression recognition for dyslexia, autism, and other disorders in specially-abled children, and various other applications could increase the use of affective computing technology. Life sciences and healthcare are prognosticated to showcase a promising rise in the demand for affective computing.

Algorithmic fairness has been studied mostly in a static setting where the implicit assumptions are that the frequencies of historically made decisions do not impact the problem structure in subsequent future. However, for example, the capability to pay back a loan for people in a certain group might depend on historically how frequently that group has been approved loan applications. If banks keep rejecting loan applications to people in a disadvantaged group, it could create a feedback loop and further damage the chance of getting loans for people in that group. This challenge has been noted in several recent works but is under-explored in a more generic sequential learning setting. In this paper, we formulate this delayed and long-term impact of actions within the context of multi-armed bandits (MAB). We generalize the classical bandit setting to encode the dependency of this action "bias" due to the history of the learning. Our goal is to learn to maximize the collected utilities over time while satisfying fairness constraints imposed over arms' utilities, which again depend on the decision they have received. We propose an algorithm that achieves a regret of $\tilde{\mathcal{O}}(KT^{2/3})$ and show a matching regret lower bound of $\Omega(KT^{2/3})$, where $K$ is the number of arms and $T$ denotes the learning horizon. Our results complement the bandit literature by adding techniques to deal with actions with long-term impacts and have implications in designing fair algorithms.

An open question in the Deep Learning community is why neural networks trained with Gradient Descent generalize well on real datasets even though they are capable of fitting random data. We propose an approach to answering this question based on a hypothesis about the dynamics of gradient descent that we call Coherent Gradients: Gradients from similar examples are similar and so the overall gradient is stronger in certain directions where these reinforce each other. Thus changes to the network parameters during training are biased towards those that (locally) simultaneously benefit many examples when such similarity exists. We support this hypothesis with heuristic arguments and perturbative experiments and outline how this can explain several common empirical observations about Deep Learning. Furthermore, our analysis is not just descriptive, but prescriptive. It suggests a natural modification to gradient descent that can greatly reduce overfitting.

Generative Adversarial Networks (GANs) are a revolutionary class of Deep Neural Networks (DNNs) that have been successfully used to generate realistic images, music, text, and other data. However, it is well known that GAN training can be notoriously resource-intensive and presents many challenges. Further, a potential weakness in GANs is that discriminator DNNs typically provide only one value (loss) of corrective feedback to generator DNNs (namely, the discriminator's assessment of the generated example). By contrast, we propose a new class of GAN we refer to as LogicGAN, that leverages recent advances in (logic-based) explainable AI (xAI) systems to provide a "richer" form of corrective feedback from discriminators to generators. Specifically, we modify the gradient descent process using xAI systems that specify the reason as to why the discriminator made the classification it did, thus providing the richer corrective feedback that helps the generator to better fool the discriminator. Using our approach, we show that LogicGANs learn much faster on MNIST data, achieving an improvement in data efficiency of 45% in single and 12.73% in multi-class setting over standard GANs while maintaining the same quality as measured by Fr\'echet Inception Distance. Further, we argue that LogicGAN enables users greater control over how models learn than standard GAN systems.

We propose a novel biologically-plausible solution to the credit assignment problem, being motivated by observations in the ventral visual pathway and trained deep neural networks. In both, representations of objects in the same category become progressively more similar, while objects belonging to different categories becomes less similar. We use this observation to motivate a layer-specific learning goal in a deep network: each layer aims to learn a representational similarity matrix that interpolates between previous and later layers. We formulate this idea using a supervised deep similarity matching cost function and derive from it deep neural networks with feedforward, lateral and feedback connections, and neurons that exhibit biologically-plausible Hebbian and anti-Hebbian plasticity. Supervised deep similarity matching can be interpreted as an energy-based learning algorithm, but with significant differences from others in how a contrastive function is constructed.

In a fast paced world where people desire more in less, Ted Talks are evolving the landscape of learning and spreading education and awareness among people who need it. This platform of education is transforming lectures into interesting interactions consuming less time as several professionals are unable to attend day-long conferences to educate and update themselves. Moreover, in terms of technology or particularly artificial intelligence (AI), the introduction of TED Talks is also beneficial with regard to money owing to its free availability online. Presenters, who are passionate technology experts, take on the stage and speak with such energy and momentum where their enthusiasm contagiously boosts up youngsters. Therefore, here we have brought you the top AI Ted Talks that will elevate your reasoning and education about the technology.