Distributional (or distribution-valued) data are a new type of data arising from several sources and are considered as realizations of distributional variables. A new set of fuzzy c-means algorithms for data described by distributional variables is proposed. The algorithms use the $L2$ Wasserstein distance between distributions as dissimilarity measures. Beside the extension of the fuzzy c-means algorithm for distributional data, and considering a decomposition of the squared $L2$ Wasserstein distance, we propose a set of algorithms using different automatic way to compute the weights associated with the variables as well as with their components, globally or cluster-wise. The relevance weights are computed in the clustering process introducing product-to-one constraints. The relevance weights induce adaptive distances expressing the importance of each variable or of each component in the clustering process, acting also as a variable selection method in clustering. We have tested the proposed algorithms on artificial and real-world data. Results confirm that the proposed methods are able to better take into account the cluster structure of the data with respect to the standard fuzzy c-means, with non-adaptive distances.

Gaussian graphical models (GGMs) are probabilistic tools of choice for analyzing conditional dependencies between variables in complex systems. Finding changepoints in the structural evolution of a GGM is therefore essential to detecting anomalies in the underlying system modeled by the GGM. In order to detect structural anomalies in a GGM, we consider the problem of estimating changes in the precision matrix of the corresponding Gaussian distribution. We take a two-step approach to solving this problem:- (i) estimating a background precision matrix using system observations from the past without any anomalies, and (ii) estimating a foreground precision matrix using a sliding temporal window during anomaly monitoring. Our primary contribution is in estimating the foreground precision using a novel contrastive inverse covariance estimation procedure. In order to accurately learn only the structural changes to the GGM, we maximize a penalized log-likelihood where the penalty is the $l_1$ norm of difference between the foreground precision being estimated and the already learned background precision. We modify the alternating direction method of multipliers (ADMM) algorithm for sparse inverse covariance estimation to perform contrastive estimation of the foreground precision matrix. Our results on simulated GGM data show significant improvement in precision and recall for detecting structural changes to the GGM, compared to a non-contrastive sliding window baseline.

Faced with massive data, is it possible to trade off (statistical) risk, and (computational) space and time? This challenge lies at the heart of large-scale machine learning. Using k-means clustering as a prototypical unsupervised learning problem, we show how we can strategically summarize the data (control space) in order to trade off risk and time when data is generated by a probabilistic model. Our summarization is based on coreset constructions from computational geometry. We also develop an algorithm, TRAM, to navigate the space/time/data/risk tradeoff in practice. In particular, we show that for a fixed risk (or data size), as the data size increases (resp. risk increases) the running time of TRAM decreases. Our extensive experiments on real data sets demonstrate the existence and practical utility of such tradeoffs, not only for k-means but also for Gaussian Mixture Models.

We propose a novel sparse tensor decomposition method, namely Tensor Truncated Power (TTP) method, that incorporates variable selection into the estimation of decomposition components. The sparsity is achieved via an efficient truncation step embedded in the tensor power iteration. Our method applies to a broad family of high dimensional latent variable models, including high dimensional Gaussian mixture and mixtures of sparse regressions. A thorough theoretical investigation is further conducted. In particular, we show that the final decomposition estimator is guaranteed to achieve a local statistical rate, and further strengthen it to the global statistical rate by introducing a proper initialization procedure. In high dimensional regimes, the obtained statistical rate significantly improves those shown in the existing non-sparse decomposition methods. The empirical advantages of TTP are confirmed in extensive simulated results and two real applications of click-through rate prediction and high-dimensional gene clustering.

In this paper, we extend graph-based identification methods by allowing background knowledge in the form of non-zero parameter values. Such information could be obtained, for example, from a previously conducted randomized experiment, from substantive understanding of the domain, or even an identification technique. To incorporate such information systematically, we propose the addition of auxiliary variables to the model, which are constructed so that certain paths will be conveniently cancelled. This cancellation allows the auxiliary variables to help conventional methods of identification (e.g., single-door criterion, instrumental variables, half-trek criterion), as well as model testing (e.g., d-separation, over-identification). Moreover, by iteratively alternating steps of identification and adding auxiliary variables, we can improve the power of existing identification methods via a bootstrapping approach that does not require external knowledge. We operationalize this method for simple instrumental sets (a generalization of instrumental variables) and show that the resulting method is able to identify at least as many models as the most general identification method for linear systems known to date. We further discuss the application of auxiliary variables to the tasks of model testing and z-identification.

In this paper, we proposed a new approximate heuristic search algorithm: Cascading A*, which is a two-phrase algorithm combining A* and IDA* by a new concept "envelope ball". The new algorithm CA* is efficient, able to generate approximate solution and any-time solution, and parallel friendly.

Why is statistical inference and machine learning approaches important for analysing Big Data? To answer this question, I want to draw your attention to the world's largest coral reef system, and one of Australia's biggest natural wonders, the Great Barrier Reef. The Great Barrier Reef is composed of over 2900 reefs and 900 islands, spanning over 2300km, and is one of the most diverse ecosystems on the Earth. However, because of its large size, monitoring and predicting different trends in the reef is really difficult. For example, here at QUT we're using machine learning approaches to design robots to seek out and control the damaging crown-of-thorns starfish. In this course we show you how to apply certain predictive analysis, dimension reduction, clustering, and machine learning techniques to analyse big data and make informed decisions.

A "robot revolution" will transform the global economy over the next 20 years, cutting the costs of doing business but exacerbating social inequality, as machines take over everything from caring for the elderly to flipping burgers, according to a new study. As well as robots performing manual jobs, such as hoovering the living room or assembling machine parts, the development of artificial intelligence means computers are increasingly able to "think", performing analytical tasks once seen as requiring human judgment. In a 300-page report, revealed exclusively to the Guardian, analysts from investment bank Bank of America Merrill Lynch draw on the latest research to outline the impact of what they regard as a fourth industrial revolution, after steam, mass production and electronics. "We are facing a paradigm shift which will change the way we live and work," the authors say. "The pace of disruptive technological innovation has gone from linear to parabolic in recent years. Penetration of robots and artificial intelligence has hit every industry sector, and has become an integral part of our daily lives."

Antwood is the alias of Tristan Douglas, a producer, microbiologist, and all-around deep thinker hailing from Nanaimo, British Columbia. Some might remember his EP Work Focus from last year, an under-the-radar gem put out by net label B.YRSLF Division. It's a slice of footwork that's been pummeled and fractured into something that's neither here nor there, which is probably why it grabbed the attention of Planet Mu boss Mike Paradinas. Around that time, Douglas was still recording under the name Margaret Antwood, an admittedly lazy spoonerism on internationally-celebrated poet and novelist Margaret Atwood. "I thought it'd be funny to take some figure that's barely known in the public consciousness and do like a really crappy pun of her name," he says.

In 2014, Facebook acquired WhatsApp for 19 billion. That astronomical number set off waves of speculation as to what value Facebook could possibly see in a company with just 55 employees and roughly 20 million in revenue, although it had 500 million users. At last week's F8 conference, that vision became a lot clearer, and it's big. Chatbots will cause a near-term disruption in how businesses interact with consumers, and a long term paradigm shift in how people will interact with machines. The easiest way to see why chatbots will make a near term impact on everyday consumers is by comparing a modern day customer support call to a chatbot experience.