In the third issue of this monthly digest series you can find out how Microsoft is bringing AI to the visually impaired, how to colorize your grayscale images, why a Google car caused a crash for the first time, and much more. Last Thursday, Microsoft showed off its Seeing AI app for the first time. It's still under development, but it looks extremely promising. Using a smartphone camera or a pair of camera-equipped smart glasses, the Seeing AI app can identify things in your environment--people, objects, and even emotions--to provide important context for what's going on around you. By a swipe of hand, the user can instruct the app to take a snapshot of the current visual scene and run it through image recognition software.

Understanding the behaviour of heuristic search methods is a challenge. This even holds for simple local search methods such as 2-OPT for the Traveling Salesperson problem. In this paper, we present a general framework that is able to construct a diverse set of instances that are hard or easy for a given search heuristic. Such a diverse set is obtained by using an evolutionary algorithm for constructing hard or easy instances that are diverse with respect to different features of the underlying problem. Examining the constructed instance sets, we show that many combinations of two or three features give a good classification of the TSP instances in terms of whether they are hard to be solved by 2-OPT.

Research on manifold learning within a density ridge estimation framework has shown great potential in recent work for both estimation and de-noising of manifolds, building on the intuitive and well-defined notion of principal curves and surfaces. However, the problem of unwrapping or unfolding manifolds has received relatively little attention within the density ridge approach, despite being an integral part of manifold learning in general. This paper proposes two novel algorithms for unwrapping manifolds based on estimated principal curves and surfaces for one- and multi-dimensional manifolds respectively. The methods of unwrapping are founded in the realization that both principal curves and principal surfaces will have inherent local maxima of the probability density function. Following this observation, coordinate systems that follow the shape of the manifold can be computed by following the integral curves of the gradient flow of a kernel density estimate on the manifold. Furthermore, since integral curves of the gradient flow of a kernel density estimate is inherently local, we propose to stitch together local coordinate systems using parallel transport along the manifold. We provide numerical experiments on both real and synthetic data that illustrates clear and intuitive unwrapping results comparable to state-of-the-art manifold learning algorithms.

This paper describes the data collection effort that is part of the project Sprekend Nederland (The Netherlands Talking), and discusses its potential use in Automatic Accent Location. We define Automatic Accent Location as the task to describe the accent of a speaker in terms of the location of the speaker and its history. We discuss possible ways of describing accent location, the consequence these have for the task of automatic accent location, and potential evaluation metrics.

We study the problem of learning sparse structure changes between two Markov networks $P$ and $Q$. Rather than fitting two Markov networks separately to two sets of data and figuring out their differences, a recent work proposed to learn changes \emph{directly} via estimating the ratio between two Markov network models. In this paper, we give sufficient conditions for \emph{successful change detection} with respect to the sample size $n_p, n_q$, the dimension of data $m$, and the number of changed edges $d$. When using an unbounded density ratio model we prove that the true sparse changes can be consistently identified for $n_p = \Omega(d^2 \log \frac{m^2+m}{2})$ and $n_q = \Omega({n_p^2})$, with an exponentially decaying upper-bound on learning error. Such sample complexity can be improved to $\min(n_p, n_q) = \Omega(d^2 \log \frac{m^2+m}{2})$ when the boundedness of the density ratio model is assumed. Our theoretical guarantee can be applied to a wide range of discrete/continuous Markov networks.

This paper explores and analyzes two randomized designs for robust Principal Component Analysis (PCA) employing low-dimensional data sketching. In one design, a data sketch is constructed using random column sampling followed by low dimensional embedding, while in the other, sketching is based on random column and row sampling. Both designs are shown to bring about substantial savings in complexity and memory requirements for robust subspace learning over conventional approaches that use the full scale data. A characterization of the sample and computational complexity of both designs is derived in the context of two distinct outlier models, namely, sparse and independent outlier models. The proposed randomized approach can provably recover the correct subspace with computational and sample complexity that are almost independent of the size of the data. The results of the mathematical analysis are confirmed through numerical simulations using both synthetic and real data.

Covariance Is 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Toyota's hybridized take on self-driving comes in the wake of research that suggests drivers can take about eight seconds or more to readjust and gain control of a formerly fully autonomous vehicle. The "guardian angel" system could alleviate that disconnect by activating only to avoid a collision, much like automatic braking does. Testing for this new system will soon be underway at a TRI location near Mt. Fuji in Japan, although the company also intends to run simulations at its US-based research facilities. To clear the system for eventual use in commercial vehicles, Toyota will first need to amass a trillion miles worth of road testing.

Drone deployment has become associated with rice growth research in the Philippines, as well as for spotting illegal fishpond-building, and for the early detection of approaching typhoons. The importance of drones has been accepted by Filipinos cognizant of neighboring countries investing millions of dollars on unmanned aerial vehicles (U.A.V.) for …

Vilson Vieira/TwitterGoogle's AI adds dog faces to Hieronymus Bosch's triptych "The Garden of Earthly Delights." In return, they discovered their algorithms can turn almost anything into trippy images of knights with dog heads and pig-snails. Now computer programmers across the internet are getting in on the "inceptionism" fun, after Google let their AI code run free on the internet. The open-source AI networks are available on GitHub for anyone with the know-how to download, use, and tweak. Gathered under the Twitter hashtag #deepdream, the resulting images range from amusing to deeply disturbing.