For the past few minutes I've been chatting with George Washington, and honestly he seems rather drunk. He also appears to have been hanging around with 20-somethings, because he keeps saying things like "cool, haha," and "u wanna join my army or wut?" This, of course, is not actually America's first president. It is automated, conversational artificial intelligence, known as a chatbot, created by Drunk History, a comedy TV show, and made available through the messaging program Kik. You can now chat with all sorts of bots through a number of messaging services including Kik, WeChat, Telegram, and now, Facebook Messenger. Some are simply meant to entertain, but a growing number are designed to do something useful.

A police investigation is under way after a passenger plane approaching Heathrow Airport hit what is believed to have been a drone. The British Airways flight from Geneva, with 132 passengers and five crew on board, was hit as it approached the London airport at 12:50 BST on Sunday. If confirmed, it is believed to be the first incident of its kind in the UK. BA said it would give the police "every assistance with their investigation". No arrests have been made, police said.

The US Department of Defense (DoD) wants contractors to mine your social media posts to develop new ways for the US government to infer what you're really thinking and feeling -- and to predict what you'll do next. Pentagon documents released over the last few months identify ongoing classified research in this area that the federal government plans to expand, by investing millions more dollars. The unclassified documents, which call on external scientists, institutions and companies to submit proposals for research projects, not only catalogue how far US military capabilities have come, but also reveal the Pentagon's goals: building the US intelligence community's capacity to forecast population behavior at home and abroad, especially groups involved in political activism. They throw light on the extent to which the Pentagon's classified pre-crime R&D has advanced, and how the US military intends to deploy it in operations around the world. A new Funding Opportunity Announcement document issued by the DoD's Office of Naval Research (ONR) calls for research proposals on how mining social media can provide insight on people's real thoughts, emotions and beliefs, and thereby facilitate predictions of behavior.

Gaussian process models are flexible, Bayesian non-parametric approaches to regression. Properties of multivariate Gaussians mean that they can be combined linearly in the manner of additive models and via a link function (like in generalized linear models) to handle non-Gaussian data. However, the link function formalism is restrictive, link functions are always invertible and must convert a parameter of interest to a linear combination of the underlying processes. There are many likelihoods and models where a non-linear combination is more appropriate. We term these more general models Chained Gaussian Processes: the transformation of the GPs to the likelihood parameters will not generally be invertible, and that implies that linearisation would only be possible with multiple (localized) links, i.e. a chain. We develop an approximate inference procedure for Chained GPs that is scalable and applicable to any factorized likelihood. We demonstrate the approximation on a range of likelihood functions.

Kernel mean embeddings have recently attracted the attention of the machine learning community. They map measures $\mu$ from some set $M$ to functions in a reproducing kernel Hilbert space (RKHS) with kernel $k$. The RKHS distance of two mapped measures is a semi-metric $d_k$ over $M$. We study three questions. (I) For a given kernel, what sets $M$ can be embedded? (II) When is the embedding injective over $M$ (in which case $d_k$ is a metric)? (III) How does the $d_k$-induced topology compare to other topologies on $M$? The existing machine learning literature has addressed these questions in cases where $M$ is (a subset of) the finite regular Borel measures. We unify, improve and generalise those results. Our approach naturally leads to continuous and possibly even injective embeddings of (Schwartz-) distributions, i.e., generalised measures, but the reader is free to focus on measures only. In particular, we systemise and extend various (partly known) equivalences between different notions of universal, characteristic and strictly positive definite kernels, and show that on an underlying locally compact Hausdorff space, $d_k$ metrises the weak convergence of probability measures if and only if $k$ is continuous and characteristic.

We consider the problem of learning a non-negative linear classifier with a $1$-norm of at most $k$, and a fixed threshold, under the hinge-loss. This problem generalizes the problem of learning a $k$-monotone disjunction. We prove that we can learn efficiently in this setting, at a rate which is linear in both $k$ and the size of the threshold, and that this is the best possible rate. We provide an efficient online learning algorithm that achieves the optimal rate, and show that in the batch case, empirical risk minimization achieves this rate as well. The rates we show are tighter than the uniform convergence rate, which grows with $k^2$.

A function $f: \mathbb{R}^d \rightarrow \mathbb{R}$ is referred to as a Sparse Additive Model (SPAM), if it is of the form $f(\mathbf{x}) = \sum_{l \in \mathcal{S}}\phi_{l}(x_l)$, where $\mathcal{S} \subset [d]$, $|\mathcal{S}| \ll d$. Assuming $\phi_l$'s and $\mathcal{S}$ to be unknown, the problem of estimating $f$ from its samples has been studied extensively. In this work, we consider a generalized SPAM, allowing for second order interaction terms. For some $\mathcal{S}_1 \subset [d], \mathcal{S}_2 \subset {[d] \choose 2}$, the function $f$ is assumed to be of the form: $$f(\mathbf{x}) = \sum_{p \in \mathcal{S}_1}\phi_{p} (x_p) + \sum_{(l,l^{\prime}) \in \mathcal{S}_2}\phi_{(l,l^{\prime})} (x_{l},x_{l^{\prime}}).$$ Assuming $\phi_{p},\phi_{(l,l^{\prime})}$, $\mathcal{S}_1$ and, $\mathcal{S}_2$ to be unknown, we provide a randomized algorithm that queries $f$ and exactly recovers $\mathcal{S}_1,\mathcal{S}_2$. Consequently, this also enables us to estimate the underlying $\phi_p, \phi_{(l,l^{\prime})}$. We derive sample complexity bounds for our scheme and also extend our analysis to include the situation where the queries are corrupted with noise -- either stochastic, or arbitrary but bounded. Lastly, we provide simulation results on synthetic data, that validate our theoretical findings.

Carl Benedikt is Co-Director of the Oxford Martin Programme on Technology and Employment at the Oxford Martin School, and Economics Associate of Nuffield College, both University of Oxford. He is also a Senior Fellow of the Programme on Employment, Equity and Growth at the Institute for New Economic Thinking in Oxford, and the Department of Economic History at Lund University. His research focuses the transition of industrial nations to digital economies, and subsequent challenges for economic growth, labour markets and urban development. To secure impact for his research outside academia, Carl Benedikt is widely engaged in policy, advisory and media activities. In partnership with Citigroup, he works to help global leaders navigate the rapidly changing world economy.

LONDON – A British Airways plane struck an object believed to be a drone on Sunday as it was coming in for landing at Heathrow, Europe's busiest airport, police said. An investigation had been launched into the incident, which follows a string of near misses involving drones. The plane, an Airbus A320 with 132 passengers and five crew members on board, was on its final descent into Heathrow's Terminal Five when it was struck. "A pilot on an inbound flight into Heathrow Airport from Geneva reported to police that he believed a drone had struck the aircraft," a spokeswoman for London's Metropolitan Police said. "The flight landed at Heathrow Terminal Five safely. It transpired that an object, believed to be a drone, had struck the front of the aircraft".

Two more near-misses between drones and passenger planes at UK airports have been reported by aviation authorities, including one a "wingspan away" from a jet landing at Heathrow. Pilots have called for a clampdown on drone use after a spate of incidents. Among the latest six to be investigated and verified by the UK Airprox Board, which monitors the threat of midair collisions, three were in the most serious bracket of risk, one involving a small light aircraft and two involving larger passenger planes. The closest calls came in late September as an Airbus A319, which typically carry up to 180 passengers, landed at Heathrow, and two days later as a turboprop commuter plane, believed to be a LoganAir flight to Scotland, left Manchester airport. The pilot at Heathrow reported a drone helicopter hovering close to his flight path, and was unable to take evasive action as the drone passed less than 30 metres away from his A319.