Simultaneous machine translation (SiMT) generates translation before reading the entire source sentence and hence it has to trade off between translation quality and latency. To fulfill the requirements of different translation quality and latency in practical applications, the previous methods usually need to train multiple SiMT models for different latency levels, resulting in large computational costs. In this paper, we propose a universal SiMT model with Mixture-of-Experts Wait-k Policy to achieve the best translation quality under arbitrary latency with only one trained model. Specifically, our method employs multi-head attention to accomplish the mixture of experts where each head is treated as a wait-k expert with its own waiting words number, and given a test latency and source inputs, the weights of the experts are accordingly adjusted to produce the best translation. Experiments on three datasets show that our method outperforms all the strong baselines under different latency, including the state-of-the-art adaptive policy.

Ultrasonic methods have great potential applications to detect and characterize defects in multi-layered bonded composites. However, it remains challenging to quantitatively reconstruct defects, such as disbonds and kissing bonds, that influence the integrity of adhesive bonds and seriously reduce the strength of assemblies. In this work, an ultrasonic method based on the supervised fully convolutional network (FCN) is proposed to quantitatively reconstruct defects hidden in multi-layered bonded composites. In the training process of this method, an FCN establishes a non-linear mapping from measured ultrasonic data to the corresponding velocity models of multi-layered bonded composites. In the predicting process, the trained network obtained from the training process is used to directly reconstruct the velocity models from the new measured ultrasonic data of adhesively bonded composites. The presented FCN-based inversion method can automatically extract useful features in multi-layered composites. Although this method is computationally expensive in the training process, the prediction itself in the online phase takes only seconds. The numerical results show that the FCN-based ultrasonic inversion method is capable to accurately reconstruct ultrasonic velocity models of the high contrast defects, which has great potential for online detection of adhesively bonded composites.

In a paper published by Nature Communication's Scientific Reports, a team of chemists from Surrey built a machine learning model based on the information from the DrugAge database to predict whether a compound can extend the life of Caenorhabditis elegans – a translucent worm that shares a similar metabolism to humans. The worm's shorter lifespan gave the researchers the opportunity to see the impact of the chemical compounds. "Ageing is increasingly being recognised as a set of diseases in modern medicine, and we can apply the tools of the digital world, such as AI, to help slow down or protect against ageing and age-related diseases. Our study demonstrates the revolutionary ability of AI to aid the identification of compounds with anti-ageing properties." "This research shows the power and potential of AI, which is a speciality of the University of Surrey, to drive significant benefits in human health."

Steel is very much prone to get defects during the manufacturing or shipping process, and it is very difficult for large manufacturing companies to detect these defects with help of manpower. Hence, there is scope to train a machine learning or deep learning model to detect these defects. Severstal is a Russian company mainly operating in the steel and mining industry, headquartered in Cherepovets. Severstal conducted a kaggle competition by providing the data of defective steel images. This story is about my work as a response to the above Kaggle competition.

FURA Gems has announced a partnership with India-based Cognecto to improve operational efficiency, sustainability, productivity and decrease the carbon footprint of its Australian mining operation. Cognecto, which calls itself India's leading artificial intelligence-based heavy equipment monitoring company, has deployed an integrated custom-built hardware sensor and remote telemetry data protocol for FURA to share the data from its Sapphire mining operations in Queensland to company headquarters in Dubai. This collaborative effort forges a solution combining heavy equipment monitoring and analytics to empower operational visibility and control wherever and whenever, according to Cognecto. In addition, FURA employees can access real-time fleet updates via a "well-integrated, easy-to-implement, and zero-tech footprint AI platform created by Cognecto to improve operational conditions and enhances safety", it said. Operational insights for real-time tracking are delivered using a web interface, while the alerts can be relayed on any commonly used messaging platform.

The almost 40-foot 3D-printed pedestrian bridge designed by Joris Laarman and built by Dutch robotics company MX3D has been opened in Amsterdam six years after the project was launched. The bridge, which was fabricated from stainless steel rods by six-axis robotic arms equipped with welding gear, spans the Oudezijds Achterburgwal in Amsterdam's Red Light District. The almost 40-foot 3D-printed pedestrian bridge designed by Joris Laarman and built by Dutch robotics company MX3D has been opened in Amsterdam six years after the project was launched. The bridge, which was fabricated from stainless steel rods by six-axis robotic arms equipped with welding gear, spans the Oudezijds Achterburgwal in Amsterdam's Red Light District. Following four years of planning and research, the world's first 3D printed footbridge recently opened to the public in Europe.

Autonomous cargo hauling won't be limited to a handful of trucks and aircraft. As CNN reports, Yara International now expects to sail the first autonomous, fully electric cargo ship in Norway by the end of 2021. The Yara Birkeland will travel from Herøya to Brevik with only three remote control centers keeping watch over the journey. Yara first developed the concept in 2017 and had planned to set sail in 2020, but the COVID-19 pandemic delayed the trip. It's not the first crewless ship of any kind to venture forth (a Finnish ferry launched in 2018), but it is the first all-electric model.

To the naked eye, a sheet of stainless steel presents a smooth, polished, homogenous surface. The same material when viewed at 400 times magnification reveals its true jumbled structure--different crystal shapes, joined at wildly different angles. Researchers at the University of Illinois Urbana-Champaign used data from high-resolution images of stainless-steel samples to train neural networks that make predictions about how the material will behave at places where the crystals meet, when strained. John Lambros explained, when studying the properties of a material such as stainless steel, it is impossible to conduct separate experiments at such high magnifications that subject it to every conceivable parameter--every temperature, every loading angle, every amount of pressure. So we often rely on models.

Segmented hinges in the long, thin bones of fish fins are critical to the incredible mechanical properties of fins, and this design could inspire improved underwater propulsion systems, new robotic materials and even new aircraft designs. Fish fins are not simple membranes that fish flap right and left for propulsion. They probably represent one of the most elegant ways to interact with water. Fins are flexible enough to morph into a wide variety of shapes, yet they are stiff enough to push water without collapsing. The secret is in the structure: Most fish have rays – long, bony spikes that stiffen the thin membranes of collagen that make up their fins.

AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results in new emergent capabilities,and their effectiveness across so many tasks incentivizes homogenization. Homogenization provides powerful leverage but demands caution, as the defects of the foundation model are inherited by all the adapted models downstream. Despite the impending widespread deployment of foundation models, we currently lack a clear understanding of how they work, when they fail, and what they are even capable of due to their emergent properties. To tackle these questions, we believe much of the critical research on foundation models will require deep interdisciplinary collaboration commensurate with their fundamentally sociotechnical nature.