Some artificial intelligence products have become so advanced in online discussion moderation that they will no longer be confused by colloquial language, neologisms or spelling mistakes. AI is able to take on routine human tasks, but cannot fully replace human intelligence. Online discussions are abound with hate speech and off-topic comments, causing massive headaches for media companies. Legislation requires that illegal messages are removed, and users are more content if they can avoid becoming the target of inappropriate insults. The volumes of comments posted on discussion forums and below news articles can be staggering, and their proper moderation may sometimes require infeasible amounts of manpower.

At the heart of AI systems are statistical models that have no concept of social inequality, fairness, or hardships. In Cathy O'Neil's book, Weapons of Math Destruction (WMD), she points out that big data is discriminating nearly at every juncture of our society and pummeling the poor at each opportunity. Her book points to many avenues of misuse of data, but most offensive is through the use of proxies. Data statistics that are designed for one purpose but are repurposed to be used for economic or convenience sake. There are a number of examples of this.

No longer the subject of science fiction, Artificial Intelligence (AI) is profoundly transforming our daily lives. While computers have been mimicking human intelligence already for some decades using logic and if-then kind of rules, massive increases in computational power are now facilitating the creation of'deep learning' machines i.e. algorithms that permit software to train itself to recognize patterns and perform tasks, like speech and image recognition, through exposure to vast amounts of data. These deep learning algorithms are everywhere, shaping our preferences and behaviour. Facebook uses a set of algorithms to tailor what news stories an individual user sees and in what order. Bot activity on Twitter last year suppressed a protest against Mexico's now-president by overloading the hashtag used to organize the event.

Artificial intelligence is everywhere: your smartphone, on streaming platforms such as Spotify and Netflix and even in some smart home appliances. But can the technology, which has seemingly caught the attention of most VCs across the world, be used in the realm of healthcare to drive efficiency and optimise patient outcomes? We take a look at some of the UK's most promising companies using AI to transform the healthcare space. No list of this kind would be complete without a mention of DeepMind, a British artificial intelligence company founded in 2010 and acquired by tech giant Google for a reported £400m four years later. DeepMind Health is leveraging machine learning technology – a form of AI – to boost the medical research field.

March 9, 2017 --Physicist Stephen Hawking may be a proponent of artificial intelligence, but he has also been outspoken about the potential challenges it creates. In a recent interview, he sounded a similar tone, and offered a solution that conservatives my find hard to accept. Speaking to The Times of London to commemorate being awarded the Honorary Freedom of the City of London, a title that was conferred on him on Monday, Professor Hawking expressed optimism for the future. He added, however, that he is concerned about artificial intelligence (AI), as well as other global threats. "We need to be quicker to identify such threats and act before they get out of control," Hawking said.

This paper proposes an innovative method for segmentation of skin lesions in dermoscopy images developed by the authors, based on fuzzy classification of pixels and histogram thresholding.

Variational inference using the reparameterization trick has enabled large-scale approximate Bayesian inference in complex probabilistic models, leveraging stochastic optimization to sidestep intractable expectations. The reparameterization trick is applicable when we can simulate a random variable by applying a differentiable deterministic function on an auxiliary random variable whose distribution is fixed. For many distributions of interest (such as the gamma or Dirichlet), simulation of random variables relies on acceptance-rejection sampling. The discontinuity introduced by the accept-reject step means that standard reparameterization tricks are not applicable. We propose a new method that lets us leverage reparameterization gradients even when variables are outputs of a acceptance-rejection sampling algorithm. Our approach enables reparameterization on a larger class of variational distributions. In several studies of real and synthetic data, we show that the variance of the estimator of the gradient is significantly lower than other state-of-the-art methods. This leads to faster convergence of stochastic gradient variational inference.

The time to converge to the steady state of a finite Markov chain can be greatly reduced by a lifting operation, which creates a new Markov chain on an expanded state space. For a class of quadratic objectives, we show an analogous behavior where a distributed ADMM algorithm can be seen as a lifting of Gradient Descent algorithm. This provides a deep insight for its faster convergence rate under optimal parameter tuning. We conjecture that this gain is always present, as opposed to the lifting of a Markov chain which sometimes only provides a marginal speedup.

User preference integration is of great importance in multi-objective optimization, in particular in many objective optimization. Preferences have long been considered in traditional multicriteria decision making (MCDM) which is based on mathematical programming. Recently, it is integrated in multi-objective metaheuristics (MOMH), resulting in focus on preferred parts of the Pareto front instead of the whole Pareto front. The number of publications on preference-based multi-objective metaheuristics has increased rapidly over the past decades. There already exist various preference handling methods and MOMH methods, which have been combined in diverse ways. This article proposes to use the Web Ontology Language (OWL) to model and systematize the results developed in this field. A review of the existing work is provided, based on which an ontology is built and instantiated with state-of-the-art results. The OWL ontology is made public and open to future extension. Moreover, the usage of the ontology is exemplified for different use-cases, including querying for methods that match an engineering application, bibliometric analysis, checking existence of combinations of preference models and MOMH techniques, and discovering opportunities for new research and open research questions.

Anyone who's ever flown a drone of any sort will tell you that sooner or later, you're going to crash it. The question is how exactly you will go about doing this, and how much of the drone will be functional after it's happened. Most flying animals somewhat frustratingly don't have this problem: Birds and insects run into things occasionally (or all the time, for small bugs), and just shrug it off and keep on going, thanks to their biological design, which includes both stiffness and flexibility. Now roboticists at the EPFL, in Lausanne, Switzerland, are relying on these same qualities to design a highly resilient quadrotor that's impressively difficult to destroy. There are three primary strategies for designing drones with impact resistance.