When Enrico Fermi decided to leave Benito Mussolini's Italy and emigrate to the United States, he changed the global balance of power. After arriving in the US, Fermi led the world's first self-sustaining nuclear reaction at the University of Chicago and played an indispensable role in the Manhattan Project, which led to the end of World War II in the Pacific and laid the groundwork for a new world order and America's prominent role. So it is not surprising that some Americans think the same should be true with AI. Emigrant AI researchers like Geoff Hinton, Yann LeCun, Yoshua Bengio, Andrew Ng, and Fei-Fei Li are the Enrico Fermis of AI and should secure an American (and Canadian) hegemony in AI. Indeed, the US and Canada have 100 percent of the top 10 AI researchers, and 68 percent of the world's best 1,000 or so researchers.

President and co-founder of US-based technology firm NVIDIA Corporation Jensen Huang Monday said artificial intelligence (AI) can be used to democratise innovation by commoditising software. He was speaking at the fourth edition of the NITI Lecture Series, which among others was attended by Prime Minister Narendra Modi. "At the 4th #NITILecture, Mr Huang of @nvidia explained how AI's power like in automating the process of automation itself. How it can write code for complex use-cases by learning from examples. "The panel discussion during the 4th #NITILecture emphasised how since #AI is heavily context dependent, it is important to evolve India-specific nuanced & inclusive solutions.

ONCE THE STUFF OF APOCALYPTIC SCI-FI tales, killer robots capable of choosing and taking out our nation's enemies are now within reach--if companies and the Pentagon decide to go that far. Defense officials have so far stopped short of developing Lethal Autonomous Weapons Systems (the government's official term), which could theoretically strike without a human order as easily as Facebook can tag friends in your photos without your say-so. But the A.I.-driven technology that could form the basis for such attacks is well underway. Project Maven, the Pentagon's most high-profile A.I. initiative, aims to use machine-learning algorithms to identify terrorist targets from drone footage, assisting military efforts to combat ISIS (more than 20 tech and defense contractors are reportedly involved, though they have not all been publicly named). Although supporting war efforts is nothing new for the defense industry, the Pentagon has increasingly looked to Silicon Valley for expertise in A.I. and facial recognition.

It was seven minutes to ten o'clock in the morning, and it was the only good thing that had happened." If you get the feeling that these sentences could have been better structured, it's simply because these seemingly disparate, literary threads have been stitched into a novel by an algorithm. That's also why the human author of this novel, Ross Godwin, calls himself'writer of writers'. He is an artist and creative technologist at Google, and also a former Obama administration ghostwriter. In March 2017, Godwin fitted a Cadillac car with a surveillance camera, global positioning system (GPS) unit, microphone and clock, and connected these devices to a portable artificial intelligence (AI) writing machine that fed on these input data in real-time.

Sorry kids, neither you nor robots can be trusted. The US Department of Transportation's National Highway Traffic Safety Administration (NHTSA) has told Transdev North America to stop hauling children around in its EZ10 Generation II autonomous shuttles in Babcock Ranch -- a community in Southwest Florida. It seems that the NHTSA has some safety concerns. "Using a non-compliant test vehicle to transport children is irresponsible, inappropriate, and in direct violation of the terms of Transdev's approved test project," NHTSA Deputy Administrator Heidi King said in a press release. Earlier this March, NHTSA approved Transdev to test and demonstrate its autonomous shuttles.

Unmanned Aerial Vehicles (UAVs) have recently rapidly grown to facilitate a wide range of innovative applications that can fundamentally change the way cyber-physical systems (CPSs) are designed. CPSs are a modern generation of systems with synergic cooperation between computational and physical potentials that can interact with humans through several new mechanisms. The main advantages of using UAVs in CPS application is their exceptional features, including their mobility, dynamism, effortless deployment, adaptive altitude, agility, adjustability, and effective appraisal of real-world functions anytime and anywhere. Furthermore, from the technology perspective, UAVs are predicted to be a vital element of the development of advanced CPSs. Therefore, in this survey, we aim to pinpoint the most fundamental and important design challenges of multi-UAV systems for CPS applications. We highlight key and versatile aspects that span the coverage and tracking of targets and infrastructure objects, energy-efficient navigation, and image analysis using machine learning for fine-grained CPS applications. Key prototypes and testbeds are also investigated to show how these practical technologies can facilitate CPS applications. We present and propose state-of-the-art algorithms to address design challenges with both quantitative and qualitative methods and map these challenges with important CPS applications to draw insightful conclusions on the challenges of each application. Finally, we summarize potential new directions and ideas that could shape future research in these areas.

Near-future large galaxy surveys will encounter blended galaxy images at a fraction of up to 50% in the densest regions of the universe. Current deblending techniques may segment the foreground galaxy while leaving missing pixel intensities in the background galaxy flux. The problem is compounded by the diffuse nature of galaxies in their outer regions, making segmentation significantly more difficult than in traditional object segmentation applications. We propose a novel branched generative adversarial network (GAN) to deblend overlapping galaxies, where the two branches produce images of the two deblended galaxies. We show that generative models are a powerful engine for deblending given their innate ability to infill missing pixel values occluded by the superposition. We maintain high peak signal-to-noise ratio and structural similarity scores with respect to ground truth images upon deblending. Our model also predicts near-instantaneously, making it a natural choice for the immense quantities of data soon to be created by large surveys such as LSST, Euclid and WFIRST.

Despite the great success achieved in machine learning (ML), adversarial examples have caused concerns with regards to its trustworthiness: A small perturbation of an input results in an arbitrary failure of an otherwise seemingly well-trained ML model. While studies are being conducted to discover the intrinsic properties of adversarial examples, such as their transferability and universality, there is insufficient theoretic analysis to help understand the phenomenon in a way that can influence the design process of ML experiments. In this paper, we deduce an information-theoretic model which explains adversarial attacks as the abuse of feature redundancies in ML algorithms. We prove that feature redundancy is a necessary condition for the existence of adversarial examples. Our model helps to explain some major questions raised in many anecdotal studies on adversarial examples. Our theory is backed up by empirical measurements of the information content of benign and adversarial examples on both image and text datasets. Our measurements show that typical adversarial examples introduce just enough redundancy to overflow the decision making of an ML model trained on corresponding benign examples. We conclude with actionable recommendations to improve the robustness of machine learners against adversarial examples.

Indicators of Compromise (IOCs) are artifacts observed on a network or in an operating system that can be utilized to indicate a computer intrusion and detect cyber-attacks in an early stage. Thus, they exert an important role in the field of cybersecurity. However, state-of-the-art IOCs detection systems rely heavily on hand-crafted features with expert knowledge of cybersecurity, and require a large amount of supervised training corpora to train an IOC classifier. In this paper, we propose using a neural-based sequence labelling model to identify IOCs automatically from reports on cybersecurity without expert knowledge of cybersecurity. Our work is the first to apply an end-to-end sequence labelling to the task in IOCs identification. By using an attention mechanism and several token spelling features, we find that the proposed model is capable of identifying the low frequency IOCs from long sentences contained in cybersecurity reports. Experiments show that the proposed model outperforms other sequence labelling models, achieving over 88% average F1-score.

Inferring missing links in knowledge graphs (KG) has attracted a lot of attention from the research community. In this paper, we tackle a practical query answering task involving predicting the relation of a given entity pair. We frame this prediction problem as an inference problem in a probabilistic graphical model and aim at resolving it from a variational inference perspective. In order to model the relation between the query entity pair, we assume that there exists an underlying latent variable (paths connecting two nodes) in the KG, which carries the equivalent semantics of their relations. However, due to the intractability of connections in large KGs, we propose to use variation inference to maximize the evidence lower bound. More specifically, our framework (\textsc{Diva}) is composed of three modules, i.e. a posterior approximator, a prior (path finder), and a likelihood (path reasoner). By using variational inference, we are able to incorporate them closely into a unified architecture and jointly optimize them to perform KG reasoning. With active interactions among these sub-modules, \textsc{Diva} is better at handling noise and coping with more complex reasoning scenarios. In order to evaluate our method, we conduct the experiment of the link prediction task on multiple datasets and achieve state-of-the-art performances on both datasets.