Physics simulators are essential in science and engineering, enabling the analysis, control, and design of complex systems. In experimental sciences, they are increasingly used to automate experimental design, often via combinatorial search and optimization. However, as the setups grow more complex, the computational cost of traditional, CPU-based simulators becomes a major limitation. Here, we show how neural surrogate models can significantly reduce reliance on such slow simulators while preserving accuracy. Taking the design of interferometric gravitational wave detectors as a representative example, we train a neural network to surrogate the gravitational wave physics simulator Finesse, which was developed by the LIGO community. Despite that small changes in physical parameters can change the output by orders of magnitudes, the model rapidly predicts the quality and feasibility of candidate designs, allowing an efficient exploration of large design spaces. Our algorithm loops between training the surrogate, inverse designing new experiments, and verifying their properties with the slow simulator for further training. Assisted by auto-differentiation and GPU parallelism, our method proposes high-quality experiments much faster than direct optimization. Solutions that our algorithm finds within hours outperform designs that take five days for the optimizer to reach. Though shown in the context of gravitational wave detectors, our framework is broadly applicable to other domains where simulator bottlenecks hinder optimization and discovery.

From the Matrix via Alien to Starship Troopers and Iron Man, exoskeletons have littered sci-fi (and adorable dog-based animations) with the promise of superhuman power for all. While medical, industrial, and military exoskeletons are advancing rapidly, nobody has managed to bring this branch of wearable tech to the masses. General Electric spent most the 1960s trying with the Hardiman exoskeleton, Samsung is still threatening to launch its GEMS (Gait Enhancing & Motivating System), and there has been a litany of failed Kickstarter projects--yet we're still waiting. Still, over the past eight months we've been made aware of three new hip-mounted, walk-assisting exoskeletons. Hypershell has three models available to preorder, and Enhanced Robotic's EX-07 promises you a new PB … soon.

Almost everyone knows HAL from '2001: A Space Odyssey,' the intelligent and, in the course of the film, increasingly sentient spaceship computer, who, with soft-spoken sangfroid, proceeds to murder the human crew that is supposed to control him. More realistic examples continue the already many decades old automation and mechanisation of labour. Today not only manual but even a range of creative tasks, like illustrating or copywriting, can be replaced by complex machines that render people not only jobless, but their skillset more or less obsolete. There is a third negative scenario that stirs up even deeper existential anxieties, not because computers become more human, but because humans become less unique, more boring, predictable, even mechanical. Using artificial neural networks, AI has the capacity to learn from its mistakes to better predict, as in the case of marketing, a customer's preferences, or, in the case of image generation programs like the popular DALL-E 2, make photorealistic or expressionistic images.

Fully cooperative, partially observable multi-agent problems are ubiquitous in the real world. In this paper, we focus on a specific subclass of coordination problems in which humans are able to discover self-explaining deviations (SEDs). SEDs are actions that deviate from the common understanding of what reasonable behavior would be in normal circumstances. They are taken with the intention of causing another agent or other agents to realize, using theory of mind, that the circumstance must be abnormal. We first motivate SED with a real world example and formalize its definition. Next, we introduce a novel algorithm, improvement maximizing self-explaining deviations (IMPROVISED), to perform SEDs. Lastly, we evaluate IMPROVISED both in an illustrative toy setting and the popular benchmark setting Hanabi, where it is the first method to produce so called finesse plays, which are regarded as one of the more iconic examples of human theory of mind.

Right now, AI music services are all the rage, and rightly so. The technology, data, and demand is there. As a producer, if you can use online tools to help inspire or improve your productions, why wouldn't you use them? And with platforms such as TikTok and YouTube, the demand to license straight-up beats and background music has never been larger. In this piece, we'll outline the AI services that you can work along side to create new formulas, sounds, and ultimately, songs.

{\alpha}{\mu} is an anytime heuristic search algorithm for incomplete information games that assumes perfect information for the opponents. {\alpha}{\mu} addresses the strategy fusion and non-locality problems encountered by Perfect Information Monte Carlo sampling. In this paper {\alpha}{\mu} is applied to the game of Bridge.

One cannot open up their computer or turn on their television for any significant amount of time without seeing or hearing about artificial intelligence. The term evokes an almost immediate emotional reaction, often with ideas of a dystopian future where the human race is no longer master of the planet. Without delving too deep into that rabbit hole, I would instead leave The Terminator and other equally bleak futures out of this particular conversation and instead focus on artificial intelligence and the law. The current application of artificial intelligence to the practice of law was a discussion topic at our most recent board of directors meeting for Loyola Law School. The discussion centered around the ability of a computer to perform a task or series of functions that had traditionally been the responsibility of a legal professional or team of professionals.

The Soli Sensor, being developed as part of a project of Google Advanced Technology and Projects group, is a low-power radar designed to use less energy and detect hand gestures on a sub-millimeter level. It operates in the 60-GHz ISM band using electromagnetic waves. The sensor detects a series of motions that are part of Soli's Virtual Tools Gestures: virtual slider, virtual button and even virtual button. The pluses for the chip are that it requires less energy, has no moving parts, can function regardless of the light conditions, and when developed further in the future, could be used in a number of products: wearables, IoT devices, phones and cars. Ivan Poupyrev, Poject Soli founder, said about the goal of the project, "The hand is the ultimate input device. How could we take this incredible capability, the finesse of human actions and the finesse of using our hand but apply it to the virtual world?"