Neural networks are increasingly employed in safety-critical domains. This has prompted interest in verifying or certifying logically encoded properties of neural networks. Prior work has largely focused on checking existential properties, wherein the goal is to check whether there exists any input that violates a given property of interest. However, neural network training is a stochastic process, and many questions arising in their analysis require probabilistic and quantitative reasoning, i.e., estimating how many inputs satisfy a given property. To this end, our paper proposes a novel and principled framework to quantitative verification of logical properties specified over neural networks. Our framework is the first to provide PAC-style soundness guarantees, in that its quantitative estimates are within a controllable and bounded error from the true count. We instantiate our algorithmic framework by building a prototype tool called NPAQ that enables checking rich properties over binarized neural networks. We show how emerging security analyses can utilize our framework in 3 concrete point applications: quantifying robustness to adversarial inputs, efficacy of trojan attacks, and fairness/bias of given neural networks.

In recent years, reinforcement learning has been successful in solving video games from Atari to Star Craft II. However, the end-to-end model-free reinforcement learning (RL) is not sample efficient and requires a significant amount of computational resources to achieve superhuman level performance. Model-free RL is also unlikely to produce human-like agents for playtesting and gameplaying AI in the development cycle of complex video games. In this paper, we present a hierarchical approach to training agents with the goal of achieving human-like style and high skill level in team sports games. While this is still work in progress, our preliminary results show that the presented approach holds promise for solving the posed multi-agent learning problem.

Designing neural architectures for edge devices is subject to constraints of accuracy, inference latency, and computational cost. Traditionally, researchers manually craft deep neural networks to meet the needs of mobile devices. Neural Architecture Search (NAS) was proposed to automate the neural architecture design without requiring extensive domain expertise and significant manual efforts. Recent works utilized NAS to design mobile models by taking into account hardware constraints and achieved state-of-the-art accuracy with fewer parameters and less computational cost measured in Multiply-accumulates (MACs). To find highly compact neural architectures, existing works relies on predefined cells and directly applying width multiplier, which may potentially limit the model flexibility, reduce the useful feature map information, and cause accuracy drop. To conquer this issue, we propose GRAM(GRAph propagation as Meta-knowledge) that adopts fine-grained (node-wise) search method and accumulates the knowledge learned in updates into a meta-graph. As a result, GRAM can enable more flexible search space and achieve higher search efficiency. Without the constraints of predefined cell or blocks, we propose a new structure-level pruning method to remove redundant operations in neural architectures. SwiftNet, which is a set of models discovered by GRAM, outperforms MobileNet-V2 by 2.15x higher accuracy density and 2.42x faster with similar accuracy. Compared with FBNet, SwiftNet reduces the search cost by 26x and achieves 2.35x higher accuracy density and 1.47x speedup while preserving similar accuracy. SwiftNetcan obtain 63.28% top-1 accuracy on ImageNet-1K with only 53M MACs and 2.07M parameters. The corresponding inference latency is only 19.09 ms on Google Pixel 1.

A remarkable feature of human beings is their capacity for creative behaviour, referring to their ability to react to problems in ways that are novel, surprising, and useful. Transformational creativity is a form of creativity where the creative behaviour is induced by a transformation of the actor's conceptual space, that is, the representational system with which the actor interprets its environment. In this report, we focus on ways of adapting systems of learned representations as they switch from performing one task to performing another. We describe an experimental comparison of multiple strategies for adaptation of learned features, and evaluate how effectively each of these strategies realizes the adaptation, in terms of the amount of training, and in terms of their ability to cope with restricted availability of training data. We show, among other things, that across handwritten digits, natural images, and classical music, adaptive strategies are systematically more effective than a baseline method that starts learning from scratch.

An engineer working for Japanese carmaker Nissan has built a robot to help farmers reduce the use of herbicides and pesticides on their rice crops. The compact robot, called Aigamo, is designed to mimic the natural use of ducks that paddle around in flooded paddy fields. Ducks have been used as natural weed repellents for centuries to tear them up and feed on insects, with their manure even acting as an additional fertiliser. As it glides through the water, two mechanisms on the bottom muddy the water to prevent weeds from getting enough sunlight to grow. The technique was used in the late 20th century with live ducks, called'aigamo,' which would paddle the water with the same results and eat any insects they found along the way.

A futuristic 3D-printed house that lets guests'experience Mars on Earth' will soon offer you the chance to experience what an interplanetary vacation of the future may be like, its creators say. Nestled in the woods of upstate New York along the Hudson River, Tera will be hired out to holiday-makers hoping to experience what sustainable life could be like on Mars. 'Tera' is the brainchild of AI SpaceFactory, a New York City design agency that was awarded $500,000 (£386,000) earlier this year for winning NASA'S 3D-Printed Habitat Challenge with its previous'Marsha' habitat. Each stay will be used to fund the mission of the firm behind its design, which hopes to research and develop the renewable and sustainable technologies of the future. This technology will be used both here on Earth and, they say, will be one day form the basis of a sustainable colony on the red planet.

DUBAI, UNITED ARAB EMIRATES - U.S. Secretary of State Mike Pompeo held talks Monday with the Saudi king and crown prince about countering the military threat from Iran by building a broad, global coalition that includes Asian and European countries. Pompeo is likely to face a tough sell in Europe and Asia, particularly from those nations still committed to the 2015 nuclear deal with Iran that President Donald Trump repudiated last year. With tensions running high in the region after Iran shot down a U.S. surveillance drone on June 20 and Trump said he aborted a retaliatory strike, Iran's naval commander warned that his forces won't hesitate to down more U.S. drones that violate its airspace. The U.S. has been building up its military presence in the Persian Gulf. The U.S. announced additional sanctions Monday on Iran aimed at pressuring the Iranian leadership into talks.

An estimated 7 million drones will be flying in the skies by 2020; Claudia Cowan reports on the new technology being developed to keep airports safe. But some people either don't care or use drones to intentionally disrupt airport operations. Last December, drone sightings at London's Gatwick Airport forced a three-day shutdown, and canceled flights left thousands of stranded passengers scrambling. No one has been arrested in the case, and this past April, investigators said it could have been an inside job. In recent months, suspected or confirmed drone activity has grounded flights in Dubai, New Zealand, Israel, and at Newark Airport in New Jersey.

Microsoft was one of the earliest companies to begin discussing and advocating for an ethical perspective on artificial intelligence. The issue began to take off at the company in 2016, when CEO Satya Nadella spoke at a developer conference about how the company viewed some of the ethical issues around AI, and later that year published an article about these issues. Nadella's primary focus was on Microsoft's orientation toward using AI to augment human capabilities and building trust into intelligent products. The next year, Microsoft's R&D head Eric Horvitz partnered with Microsoft's president and chief legal officer Brad Smith to form Aether, a cross-functional committee addressing AI and ethics in engineering and research. With these foundations laid, in 2018, Microsoft established a full-time position in AI policy and ethics.

From flying vehicles to smart buildings, this year's IoTSWC will bring the best industrial internet solutions to Barcelona. The Industrial Internet Consortium (IIC) is the leading organization promoting Industry 4.0 in the US. During the last five years, together with Fira Barcelona, it has been organizing the IoT Solutions World Congress (IoTSWC), the leading conference of industrial IoT. Dr. Richard Soley is the Executive Director of the Industrial Internet Consortium and is responsible for the vision and direction of the organization. In addition to this role, Dr. Soley is Chairman and CEO of the Object Management Group (OMG) – an international, nonprofit computer industry standards consortium -- and Executive Director of the Cloud Standards Customer Council – an end-user advocacy group.