Computer mastery of the complex board game Go has long been seen as a landmark for artificial intelligence. A mystery player causing a stir in the world of the complex strategy game Go has been revealed as an updated version of AlphaGo, the artificial-intelligence program created by Google's London-based AI firm, DeepMind. Known only by the name "Master (P)", since late December the anonymous player has beaten the world's best at Go in a string of online games, including defeating current world number one, 19-year old Ke Jie. South Korea trumpets $860-million AI fund after AlphaGo'shock' Go is regarded as the most complex board game ever invented, and famously difficult for computers to crack. But last year, AlphaGo showcased the strength of AI software when it stunned the Go world first by defeating a professional human player, Fan Hui, and then going on to beat one of the Go world's top players, Lee Sedol.

AI is going everywhere – into space, inside the human body and now, into your mouth. Ara, the world's first AI-capable toothbrush will actually teach you how to brush your teeth better. If you're a dentist, you'll probably get to keep your day job. Ara, from Kolibree, a French oral care company with offices in France and the United States, connects with an app on your smartphone to transmit data about your brushing habits. It will know exactly how you brush and whether or not it's good enough to keep dental problems at bay.

Computation of moments of transformed random variables is a problem appearing in many engineering applications. The current methods for moment transformation are mostly based on the classical quadrature rules which cannot account for the approximation errors. Our aim is to design a method for moment transformation for Gaussian random variables which accounts for the error in the numerically computed mean. We employ an instance of Bayesian quadrature, called Gaussian process quadrature (GPQ), which allows us to treat the integral itself as a random variable, where the integral variance informs about the incurred integration error. Experiments on the coordinate transformation and nonlinear filtering examples show that the proposed GPQ moment transform performs better than the classical transforms.

In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active towards a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria) it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecules for drug discovery.

Dynamic Mode Decomposition (DMD) has emerged as a powerful tool for analyzing the dynamics of non-linear systems from experimental datasets. Recently, several attempts have extended DMD to the context of low-rank approximations. This extension is of particular interest for reduced-order modeling in various applicative domains, e.g. for climate prediction, to study molecular dynamics or micro-electromechanical devices. This low-rank extension takes the form of a non-convex optimization problem. To the best of our knowledge, only sub-optimal algorithms have been proposed in the literature to compute the solution of this problem. In this paper, we prove that there exists a closed-form optimal solution to this problem and design an effective algorithm to compute it based on Singular Value Decomposition (SVD). A toy-example illustrates the gain in performance of the proposed algorithm compared to state-of-the-art techniques.

The constraint satisfaction problem (CSP) involves deciding, given a set of variables and a set of constraints on the variables, whether or not there is an assignment to the variables satisfying all of the constraints. One formulation of the CSP is as the problem of deciding, given a pair (G,H) of relational structures, whether or not there is a homomorphism from the first structure to the second structure. The CSP is in general NP-hard; a common way to restrict this problem is to fix the second structure H, so that each structure H gives rise to a problem CSP(H). The problem family CSP(H) has been studied using an algebraic approach, which links the algorithmic and complexity properties of each problem CSP(H) to a set of operations, the so-called polymorphisms of H. Certain types of polymorphisms are known to imply the polynomial-time tractability of $CSP(H)$, and others are conjectured to do so. This article systematically studies---for various classes of polymorphisms---the computational complexity of deciding whether or not a given structure H admits a polymorphism from the class. Among other results, we prove the NP-completeness of deciding a condition conjectured to characterize the tractable problems CSP(H), as well as the NP-completeness of deciding if CSP(H) has bounded width.

We consider existential rules (aka Datalog+) as a formalism for specifying ontologies. In recent years, many classes of existential rules have been exhibited for which conjunctive query (CQ) entailment is decidable. However, most of these classes cannot express transitivity of binary relations, a frequently used modelling construct. In this paper, we address the issue of whether transitivity can be safely combined with decidable classes of existential rules. First, we prove that transitivity is incompatible with one of the simplest decidable classes, namely aGRD (acyclic graph of rule dependencies), which clarifies the landscape of `finite expansion sets' of rules. Second, we show that transitivity can be safely added to linear rules (a subclass of guarded rules, which generalizes the description logic DL-Lite-R) in the case of atomic CQs, and also for general CQs if we place a minor syntactic restriction on the rule set. This is shown by means of a novel query rewriting algorithm that is specially tailored to handle transitivity rules. Third, for the identified decidable cases, we pinpoint the combined and data complexities of query entailment.

This paper presents formulae that can solve various seemingly hopeless philosophical conundrums. We discuss the simulation argument, teleportation, mind-uploading, the rationality of utilitarianism, and the ethics of exploiting artificial general intelligence. Our approach arises from combining the essential ideas of formalisms such as algorithmic probability, the universal intelligence measure, space-time-embedded intelligence, and Hutter's observer localization. We argue that such universal models can yield the ultimate solutions, but a novel research direction would be required in order to find computationally efficient approximations thereof.

What's all the fuss about the voice-activated home speaker that Amazon is due to release in the UK and Germany in late September? This gadget has been available in the US for over a year and has proven a minor hit, with sales estimates between 1.6m and 3m. But these figures belie the potential impact this kind of artificial intelligence device could have on our lives in the near future. Echo doesn't just let you switch on your music by voice command. It's the first of what will be several types of smart home appliances that work beyond simple tasks like playing music or turning on a light.

PDF: We made a fancy PDF of this post for printing and offline viewing. Note: The reason this post took three weeks to finish is that as I dug into research on Artificial Intelligence, I could not believe what I was reading. It hit me pretty quickly that what's happening in the world of AI is not just an important topic, but by far THE most important topic for our future. So I wanted to learn as much as I could about it, and once I did that, I wanted to make sure I wrote a post that really explained this whole situation and why it matters so much. Not shockingly, that became outrageously long, so I broke it into two parts. This is Part 1--Part 2 is here. We are on the edge of change comparable to the rise of human life on Earth. It seems like a pretty intense place to be standing--but then you have to remember something about what it's like to stand on a time graph: you can't see what's to your right. So here's how it actually feels to stand there: Imagine taking a time machine back to 1750--a time when the world was in a permanent power outage, long-distance communication meant either yelling loudly or firing a cannon in the air, and all transportation ran on hay. When you get there, you retrieve a dude, bring him to 2015, and then walk him around and watch him react to everything.