We present novel graph kernels for graphs with node and edge labels that have ordered neighborhoods, i.e. when neighbor nodes follow an order. Graphs with ordered neighborhoods are a natural data representation for evolving graphs where edges are created over time, which induces an order. Combining convolutional subgraph kernels and string kernels, we design new scalable algorithms for generation of explicit graph feature maps using sketching techniques. We obtain precise bounds for the approximation accuracy and computational complexity of the proposed approaches and demonstrate their applicability on real datasets. In particular, our experiments demonstrate that neighborhood ordering results in more informative features. For the special case of general graphs, i.e. graphs without ordered neighborhoods, the new graph kernels yield efficient and simple algorithms for the comparison of label distributions between graphs.

A key problem in location-based modeling and forecasting lies in identifying suitable spatial and temporal resolutions. In particular, judicious spatial partitioning can play a significant role in enhancing the performance of location-based forecasting models. In this work, we investigate two widely used tessellation strategies for partitioning city space, in the context of real-time taxi demand forecasting. Our study compares (i) Geohash tessellation, and (ii) Voronoi tessellation, using two distinct taxi demand datasets, over multiple time scales. For the purpose of comparison, we employ classical time-series tools to model the spatio-temporal demand. Our study finds that the performance of each tessellation strategy is highly dependent on the city geography, spatial distribution of the data, and the time of the day, and that neither strategy is found to perform optimally across the forecast horizon. We propose a hybrid tessellation algorithm that picks the best tessellation strategy at each instant, based on their performance in the recent past. Our hybrid algorithm is a non-stationary variant of the well-known HEDGE algorithm for choosing the best advice from multiple experts. We show that the hybrid tessellation strategy performs consistently better than either of the two strategies across the data sets considered, at multiple time scales, and with different performance metrics. We achieve an average accuracy of above 80% per km^2 for both data sets considered at 60 minute aggregation levels.

We propose a new sample-efficient methodology, called Supervised Policy Update (SPU), for deep reinforcement learning. Starting with data generated by the current policy, SPU optimizes over the proximal policy space to find a non-parameterized policy. It then solves a supervised regression problem to convert the non-parameterized policy to a parameterized policy, from which it draws new samples. There is significant flexibility in setting the labels in the supervised regression problem, with different settings corresponding to different underlying optimization problems. We develop a methodology for finding an optimal policy in the non-parameterized policy space, and show how Trust Region Policy Optimization (TRPO) and Proximal Policy Optimization (PPO) can be addressed by this methodology. In terms of sample efficiency, our experiments show SPU can outperform PPO for simulated robotic locomotion tasks.

Many iterative and non-iterative methods have been developed for inverse problems associated with Ising models. Aiming to derive an accurate non-iterative method for the inverse problems, we employ the tree-reweighted approximation. Using the tree-reweighted approximation, we can optimize the rigorous lower bound of the objective function. By solving the moment-matching and self-consistency conditions analytically, we can derive the interaction matrix as a function of the given data statistics. With this solution, we can obtain the optimal interaction matrix without iterative computation. To evaluate the accuracy of the proposed inverse formula, we compared our results to those obtained by existing inverse formulae derived with other approximations. In an experiment to reconstruct the interaction matrix, we found that the proposed formula returns the best estimates in strongly-attractive regions for various graph structures. We also performed an experiment using real-world biological data. When applied to finding the connectivity of neurons from spike train data, the proposed formula gave the closest result to that obtained by a gradient ascent algorithm, which typically requires thousands of iterations.

New research out of MIT's Media Lab is underscoring what other experts have reported or at least suspected before: facial recognition technology is subject to biases based on the data sets provided and the conditions in which algorithms are created. Joy Buolamwini, a researcher at the MIT Media Lab, recently built a dataset of 1,270 faces, using the faces of politicians, selected based on their country's rankings for gender parity (in other words, having a significant number of women in public office). Buolamwini then tested the accuracy of three facial recognition systems: those made by Microsoft, IBM, and Megvii of China. The results, which were originally reported in The New York Times, showed inaccuracies in gender identification dependent on a person's skin color. Gender was misidentified in less than one percent of lighter-skinned males; in up to seven percent of lighter-skinned females; up to 12 percent of darker-skinned males; and up to 35 percent in darker-skinner females. "Overall, male subjects were more accurately classified than female subjects replicating previous findings (Ngan et al., 2015), and lighter subjects were more accurately classified than darker individuals," Buolamwini wrote in a paper about her findings, which was co-authored by Timnit Gebru, a Microsoft researcher.

While many companies view artificial intelligence as an integral part of their future success, very few have fully embraced the technology. According to a global Accenture survey, business executives believe that within the next two years, artificial intelligence will work next to humans in their organisations as a co-worker, collaborator and trusted advisor. Accenture predicts by 2022, firms that adopt AI can boost revenues by up to 38 per cent. But rolling out AI at scale across a company is far easier said than done. And despite the lofty ambitions, very few businesses locally have taken AI beyond the experimental stage. There are still significant hurdles to be overcome when adopting AI in a significant way.

Upon closer inspection, the satellite had noticed that an area that should have been shrouded in forest, was now barren. Within hours, a call had been made to a global conservation group, who mounted a legal case against the logging companies operating in the area. That process, historically, could have taken months of observing and recording changes. What's more, in remote areas such as the Ussuri Taiga in Russia's Far East, policing illegal logging operations have historically had little impact on the extraction of timber. But thanks to artificial intelligence (AI) and satellites, the ability to observe and respond to changes has become much faster.

Several companies are working on attempting to improve the display quality for virtual reality (VR), with many concentrating on attempting to improve the resolution and refresh rate, but this does come with an attendant additional load on the processor and graphics card. Electronics company LG are trying out a different solution, one which involves artificial intelligence (AI). LG Display have been working together with Sogang University in South Korea to create an AI-powered algorithm that can reduce latency and motion blue in VR content. High latency has been shown to be one of the factors that can cause simulation sickness symptoms, along with motion blur. Higher-resolution displays can even exacerbate this problem since they require more calculations which can increase the latency. The LG technology uses an AI algorithm that converts low-resolution video into higher resolutions in real time.

MUMBAI: As job disruption intensifies due to adoption of next-gen technology, Internet of things (IoT) and artificial intelligence (AI) have emerged as the most sought-after courses for upskilling among mid-level and even senior executives. Technology professionals from all experience levels are going in for reskilling driven by digitisation, the need to stay relevant and for job security. AI and IoT are, therefore, the hottest skills to acquire for 2018, as well as the fastest growing, according to research shared exclusively with TOI by Edureka, an online learning company that offers over 100 courses in trending technologies. Even hiring data suggests new technology trends such as AI, blockchain and cyber-security will sweep the market off conventional skills. AI, which seems to be becoming a mainstay in almost every form of technology, is going to pop up in all new platforms, devices or apps more and more through 2018, with over 50% firms implementing AI in their products, says a survey by Nasscom.

A blood-thirsty new video game that encourages players to take part in a school shooting has triggered outrage online. The upcoming game, Active Shooter, lets players choose between taking on the role of a SWAT team member trying to stop an ongoing school shooting -- or the role of the shooter themselves. Those who play as the shooter will be shown a tally of the number of civilians and police officers they have managed to kill during their simulated shooting spree. Anti-gun violence charity Infer Trust has described the game as'horrendous' and in'bad taste' given the recent mass shootings in the US. Players in'Active Shooter' will be shown a tally of the number of civilians and police officers they have killed during their simulated school shooting spree'Pick your role, gear up and fight or destroy!' the description on the Active Shooter listing declares. 'Only in "Active Shooter", you will be able to pick the role of an Elite S.W.A.T team member or the actual shooter.