ZHUHAI, CHINA – China is unleashing stealth drones and pilotless aircraft fitted with AK-47 rifles onto world markets, racing to catch up to U.S. technology and adding to a fleet that has already seen combat action in the Middle East. Combat drones were among the jet fighters, missiles and other military hardware shown off this past week at Airshow China, the country's biggest aerospace industry exhibition. A delta-winged stealth drone received much attention, highlighting China's growing production of sophisticated unmanned aerial vehicles seeking to compete with the U.S. military's massive fleet. The CH-7 -- a charcoal-gray UAV unveiled at the air show -- is as long as a tennis court and has a 22-meter (72-feet) wingspan. It can fly at more than 800 kph (500 mph) and at an altitude of 13,000 meters (42,650 feet).

Recently, a series of models have been proposed to predict future movements of people. Meanwhile, dictionary coding algorithms are used to estimate the predictability limit of human mobility. Although dictionary coding is optimal, it takes long time to converge. Consequently, it is ineffective to infer predictability through dictionary coding algorithms. In this report, we illustrate this ineffectiveness on the basis of human movements in urban space.

We tackle the blackbox issue of deep neural networks in the settings of reinforcement learning (RL) where neural agents learn towards maximizing reward gains in an uncontrollable way. Such learning approach is risky when the interacting environment includes an expanse of state space because it is then almost impossible to foresee all unwanted outcomes and penalize them with negative rewards beforehand. Unlike reverse analysis of learned neural features from previous works, our proposed method \nj{tackles the blackbox issue by encouraging} an RL policy network to learn interpretable latent features through an implementation of a disentangled representation learning method. Toward this end, our method allows an RL agent to understand self-efficacy by distinguishing its influences from uncontrollable environmental factors, which closely resembles the way humans understand their scenes. Our experimental results show that the learned latent factors not only are interpretable, but also enable modeling the distribution of entire visited state space with a specific action condition. We have experimented that this characteristic of the proposed structure can lead to ex post facto governance for desired behaviors of RL agents.

Providing reinforcement learning agents with informationally rich human knowledge can dramatically improve various aspects of learning. Prior work has developed different kinds of shaping methods that enable agents to learn efficiently in complex environments. All these methods, however, tailor human guidance to agents in specialized shaping procedures, thus embodying various characteristics and advantages in different domains. In this paper, we investigate the interplay between different shaping methods for more robust learning performance. We propose an adaptive shaping algorithm which is capable of learning the most suitable shaping method in an on-line manner. Results in two classic domains verify its effectiveness from both simulated and real human studies, shedding some light on the role and impact of human factors in human-robot collaborative learning.

We propose DecaProp (Densely Connected Attention Propagation), a new densely connected neural architecture for reading comprehension (RC). There are two distinct characteristics of our model. Firstly, our model densely connects all pairwise layers of the network, modeling relationships between passage and query across all hierarchical levels. Secondly, the dense connectors in our network are learned via attention instead of standard residual skip-connectors. To this end, we propose novel Bidirectional Attention Connectors (BAC) for efficiently forging connections throughout the network. We conduct extensive experiments on four challenging RC benchmarks. Our proposed approach achieves state-of-the-art results on all four, outperforming existing baselines by up to $2.6\%-14.2\%$ in absolute F1 score.

China's Xinhua News Agency has introduced an artificial intelligence news anchor. China's Xinhua News Agency has introduced an artificial intelligence news anchor. "This is my very first day at Xinhua News Agency," says a sharply dressed artificial intelligence news anchor. "I look forward to bringing you the brand new news experiences." China's Xinhua News Agency has billed the technology as the "world's first artificial intelligence (AI) news anchor," unveiled at the World Internet Conference in China's Zhejiang province.

I've been using the OnePlus 6T for almost a month now, and I really like it. It's very much a continuation of the styling OnePlus used on the OnePlus 6 earlier in the year. That was the first to break with the metal-back styling of phones since the OnePlus 3 and opt for a glass enclosure front and rear. This looks as smart and stylish as the last model – which is very stylish, by the way – and although it's a little thicker from front to back (8.2mm instead of the OnePlus 6's 7.8mm), it feels very comfortable in the hand, not least because it's narrower than the last model. Still, it squeezes in a bigger 6.4-inch display because the screen proportions are different: This is a 19.5:9 ratio screen.

Chinese phone company Xiaomi has finally launched its phones in the UK, ready to take on companies like Apple and Samsung. As well as its Mi 8 Pro flagship phone – which is now on sale, the first time the phone is available outside of China – the company is selling a wristband and a scooter as it attempts to spread across the world. The company is already the fourth biggest phone company in the world, after Samsung, Huawei and Apple. But it has mostly sold its products in China, and is now seeking to spread into more countries. The I.F.O. is fuelled by eight electric engines, which is able to push the flying object to an estimated top speed of about 120mph.

Accenture ACN, 0.40% has invested in and formed a strategic alliance with Quantexa, a leading data analytics firm that provides technology solutions to uncover new intelligence. The minority investment made by Accenture Ventures will help Quantexa accelerate the development of its AI-powered entity resolution and network analytics technology working with Accenture Applied Intelligence. The new collaboration will help clients uncover new actionable insights across multiple industry use cases including fighting financial crime. As part of the alliance agreement, Accenture plans to build new capabilities that combine its technology and risk expertise with Quantexa's network analytics platform. The collaboration aims to develop multiple AI-enabled solutions addressing business challenges in areas including anti-money laundering, credit risk and customer insight.

The traditional hiring process is failing candidates, companies and recruiters. It revolves around human interpretation of complex data that is too susceptible to prejudices and mental shortcuts. A hiring process assisted by machine learning could eliminate systemic biases that put social status over skill. However, in recent news, algorithms have failed at that task. A resume screening algorithm developed at Amazon became biased against female applicants.