When I go to Google Photos and search my photos for'skyline', it finds me this picture of the New York skyline I took in August, without me having labelled it! When I search for'cathedral', Google's neural networks find me pictures of cathedrals & churches I've seen. But of course, neural networks aren't magic–nothing is! I recently read a paper, "Explaining and Harnessing Adversarial Examples", that helped demystify neural networks a little for me. The paper explains how to force a neural network to make really egregious mistakes. It does this by exploiting the fact that the network is simpler (more linear!) than you might expect. It's important to understand that this doesn't explain all (or even most) kinds of mistakes neural networks make. There are a lot of possible mistakes!

Google has updated the app for users in 62 countries around the world The app now uses machine learning to suggest songs and playlists Subscribers will be able to listen to music offline, and the app will automatically create an offline playlist based on what you've listened to Subscribers will be able to listen to music offline, and the app will automatically create an offline playlist based on what you've listened to The latest feature to receive the'smart' treatment, is Google Play Music, which has been updated to be more assistive. Sex with robots will be'mind blowing': Expert says androids... Instagram wants to make'Stories' more sociable: App adds... You can finally send GIFs on WhatsApp for iOS: App... Were YOU killed off by Facebook? Sex with robots will be'mind blowing': Expert says androids... Instagram wants to make'Stories' more sociable: App adds... You can finally send GIFs on WhatsApp for iOS: App... Were YOU killed off by Facebook?

In classical reinforcement learning, when exploring an environment, agents accept arbitrary short term loss for long term gain. This is infeasible for safety critical applications, such as robotics, where even a single unsafe action may cause system failure. In this paper, we address the problem of safely exploring finite Markov decision processes (MDP). We define safety in terms of an, a priori unknown, safety constraint that depends on states and actions. We aim to explore the MDP under this constraint, assuming that the unknown function satisfies regularity conditions expressed via a Gaussian process prior. We develop a novel algorithm for this task and prove that it is able to completely explore the safely reachable part of the MDP without violating the safety constraint. To achieve this, it cautiously explores safe states and actions in order to gain statistical confidence about the safety of unvisited state-action pairs from noisy observations collected while navigating the environment. Moreover, the algorithm explicitly considers reachability when exploring the MDP, ensuring that it does not get stuck in any state with no safe way out. We demonstrate our method on digital terrain models for the task of exploring an unknown map with a rover.

Predictive models are increasingly deployed for the purpose of determining access to services such as credit, insurance, and employment. Despite potential gains in productivity and efficiency, several potential problems have yet to be addressed, particularly the potential for unintentional discrimination. We present an iterative procedure, based on orthogonal projection of input attributes, for enabling interpretability of black-box predictive models. Through our iterative procedure, one can quantify the relative dependence of a black-box model on its input attributes.The relative significance of the inputs to a predictive model can then be used to assess the fairness (or discriminatory extent) of such a model.

W e propose a new framework for the analysis of low-rank tensors which lies at the intersection of spectral graph theory and signal processing. As a first step, we present a new graph based low-rank decomposition which approximates the classical low-rank SVD for matrices and multi-linear SVD for tensors. Then, building on this novel decomposition we construct a general class of convex optimization problems for approximately solving low-rank tensor inverse problems, such as tensor Robust PCA. The whole framework is named as "Multilinear Low-rank tensors on Graphs (MLRTG)". Our theoretical analysis shows: 1) MLRTG stands on the notion of approximate stationarity of multidimensional signals on graphs and 2) the approximation error depends on the eigen gaps of the graphs. W e demonstrate applications for a wide variety of 4 artificial and 12 real tensor datasets, such as EEG, FMRI, BCI, surveillance videos and hyperspectral images. Generalization of the tensor concepts to non-euclidean domain, orders of magnitude speedup, low-memory requirement and significantly enhanced performance at low SNR are the key aspects of our framework.

The ability to track a moving vehicle is of crucial importance in numerous applications. The task has often been approached by the importance sampling technique of particle filters due to its ability to model non-linear and non-Gaussian dynamics, of which a vehicle travelling on a road network is a good example. Particle filters perform poorly when observations are highly informative. In this paper, we address this problem by proposing particle filters that sample around the most recent observation. The proposal leads to an order of magnitude improvement in accuracy and efficiency over conventional particle filters, especially when observations are infrequent but low-noise.

New proposed models are often compared to state-of-the-art using statistical significance testing. Literature is scarce for classifier comparison using metrics other than accuracy. We present a survey of statistical methods that can be used for classifier comparison using precision, accounting for inter-precision correlation arising from use of same dataset. Comparisons are made using per-class precision and methods presented to test global null hypothesis of an overall model comparison. Comparisons are extended to multiple multi-class classifiers and to models using cross validation or its variants. Partial Bayesian update to precision is introduced when population prevalence of a class is known. Applications to compare deep architectures are studied.

Amazon's Spotify-rival arrives in the UK: Music Unlimited runs through AI assistant Alexa and offers ... UT's Parker Leading White House, NSF Effort to Boost Artificial Intelligence

Lower and middle-skilled roles, such as routine manual or data processing jobs, are at risk from developing AI, according to a report published by the Government Office for Science. Outlining some of the possible implications of AI, the report says new technologies such as machine learning, robotics, big data and autonomous systems could have huge implications for the economy and labour markets. It reads: "These technologies together can be seen as part of a new wave of'general purpose' digital technologies, comparable to the steam engine, and the moving assembly line, with the potential to drive significant socio-economic change." The extent and speed at which new technologies will impact the labour market is still uncertain, however. While a Deloitte study quoted by the report found that 35% of UK jobs will be affected by automation over the next 10 to 20 years, the OECD said only 10% of jobs are at risk.

VIDEO: San Jose airport unveils robots to help passengers Robots at center of China's strategy to leapfrog rivals Robots at center of China's strategy to leapfrog rivals Robots at center of China's strategy to leapfrog rivals REDWOOD CITY (KRON)--Redwood City officials approved a pilot program this week that allows robots to deliver groceries and restaurant food to homes and businesses. Starship Technologies Inc. will launch the service with 20 robots. It will begin next month and last for nine months. Each robot is designed to carry three grocery bags and use nine on-board cameras to navigate. They travel four miles per hour and will be controlled by someone when they are crossing intersections, city officials said.